Make image output lossy (use stb image writer), also significantly improves performance for jpeg

10 files changed, 324 insertions, 350 deletions

diff --git a/src/capture/capture.c b/src/capture/capture.c
index 2a4a689..bc95300 100644
--- a/src/capture/capture.c
+++ b/src/capture/capture.c
@@ -1,9 +1,9 @@
 #include "../../include/capture/capture.h"
 #include <assert.h>
 
-int gsr_capture_start(gsr_capture *cap, AVCodecContext *video_codec_context, AVFrame *frame) {
+int gsr_capture_start(gsr_capture *cap, gsr_capture_metadata *capture_metadata) {
     assert(!cap->started);
-    int res = cap->start(cap, video_codec_context, frame);
+    int res = cap->start(cap, capture_metadata);
     if(res == 0)
         cap->started = true;
 
@@ -29,9 +29,9 @@ bool gsr_capture_should_stop(gsr_capture *cap, bool *err) {
         return false;
 }
 
-int gsr_capture_capture(gsr_capture *cap, AVFrame *frame, gsr_color_conversion *color_conversion) {
+int gsr_capture_capture(gsr_capture *cap, gsr_capture_metadata *capture_metadata, gsr_color_conversion *color_conversion) {
     assert(cap->started);
-    return cap->capture(cap, frame, color_conversion);
+    return cap->capture(cap, capture_metadata, color_conversion);
 }
 
 bool gsr_capture_uses_external_image(gsr_capture *cap) {
@@ -48,6 +48,6 @@ bool gsr_capture_set_hdr_metadata(gsr_capture *cap, AVMasteringDisplayMetadata *
         return false;
 }
 
-void gsr_capture_destroy(gsr_capture *cap, AVCodecContext *video_codec_context) {
-    cap->destroy(cap, video_codec_context);
+void gsr_capture_destroy(gsr_capture *cap) {
+    cap->destroy(cap);
 }
diff --git a/src/capture/kms.c b/src/capture/kms.c
index ae0c36f..266d4e6 100644
--- a/src/capture/kms.c
+++ b/src/capture/kms.c
@@ -55,7 +55,6 @@ typedef struct {
     bool is_x11;
     gsr_cursor x11_cursor;
 
-    AVCodecContext *video_codec_context;
     bool performance_error_shown;
     bool fast_path_failed;
     bool mesa_supports_compute_only_vaapi_copy;
@@ -177,7 +176,7 @@ static vec2i rotate_capture_size_if_rotated(gsr_capture_kms *self, vec2i capture
     return capture_size;
 }
 
-static int gsr_capture_kms_start(gsr_capture *cap, AVCodecContext *video_codec_context, AVFrame *frame) {
+static int gsr_capture_kms_start(gsr_capture *cap, gsr_capture_metadata *capture_metadata) {
     gsr_capture_kms *self = cap->priv;
 
     gsr_capture_kms_create_input_texture_ids(self);
@@ -219,17 +218,14 @@ static int gsr_capture_kms_start(gsr_capture *cap, AVCodecContext *video_codec_c
     else
         self->capture_size = rotate_capture_size_if_rotated(self, monitor.size);
 
-    /* Disable vsync */
-    self->params.egl->eglSwapInterval(self->params.egl->egl_display, 0);
-
     if(self->params.output_resolution.x == 0 && self->params.output_resolution.y == 0) {
         self->params.output_resolution = self->capture_size;
-        video_codec_context->width = FFALIGN(self->capture_size.x, 2);
-        video_codec_context->height = FFALIGN(self->capture_size.y, 2);
+        capture_metadata->width = FFALIGN(self->capture_size.x, 2);
+        capture_metadata->height = FFALIGN(self->capture_size.y, 2);
     } else {
         self->params.output_resolution = scale_keep_aspect_ratio(self->capture_size, self->params.output_resolution);
-        video_codec_context->width = FFALIGN(self->params.output_resolution.x, 2);
-        video_codec_context->height = FFALIGN(self->params.output_resolution.y, 2);
+        capture_metadata->width = FFALIGN(self->params.output_resolution.x, 2);
+        capture_metadata->height = FFALIGN(self->params.output_resolution.y, 2);
     }
 
     self->fast_path_failed = self->params.egl->gpu_info.vendor == GSR_GPU_VENDOR_AMD && !gl_driver_version_greater_than(&self->params.egl->gpu_info, 24, 0, 9);
@@ -243,10 +239,6 @@ static int gsr_capture_kms_start(gsr_capture *cap, AVCodecContext *video_codec_c
 
     self->mesa_supports_compute_only_vaapi_copy = self->params.egl->gpu_info.vendor == GSR_GPU_VENDOR_AMD && gl_driver_version_greater_than(&self->params.egl->gpu_info, 24, 3, 6);
 
-    frame->width = video_codec_context->width;
-    frame->height = video_codec_context->height;
-
-    self->video_codec_context = video_codec_context;
     self->last_time_monitor_check = clock_get_monotonic_seconds();
     return 0;
 }
@@ -617,7 +609,7 @@ static void gsr_capture_kms_fail_fast_path_if_not_fast(gsr_capture_kms *self, ui
     }
 }
 
-static int gsr_capture_kms_capture(gsr_capture *cap, AVFrame *frame, gsr_color_conversion *color_conversion) {
+static int gsr_capture_kms_capture(gsr_capture *cap, gsr_capture_metadata *capture_metadata, gsr_color_conversion *color_conversion) {
     gsr_capture_kms *self = cap->priv;
 
     gsr_capture_kms_cleanup_kms_fds(self);
@@ -648,7 +640,7 @@ static int gsr_capture_kms_capture(gsr_capture *cap, AVFrame *frame, gsr_color_c
     if(drm_fd->has_hdr_metadata && self->params.hdr && hdr_metadata_is_supported_format(&drm_fd->hdr_metadata))
         gsr_kms_set_hdr_metadata(self, drm_fd);
 
-    if(!self->performance_error_shown && self->monitor_rotation != GSR_MONITOR_ROT_0 && video_codec_context_is_vaapi(self->video_codec_context) && self->params.egl->gpu_info.vendor == GSR_GPU_VENDOR_AMD) {
+    if(!self->performance_error_shown && self->monitor_rotation != GSR_MONITOR_ROT_0 && video_codec_context_is_vaapi(capture_metadata->video_codec_context) && self->params.egl->gpu_info.vendor == GSR_GPU_VENDOR_AMD) {
         self->performance_error_shown = true;
         self->fast_path_failed = true;
         fprintf(stderr, "gsr warning: gsr_capture_kms_capture: the monitor you are recording is rotated, composition will have to be used."
@@ -664,7 +656,7 @@ static int gsr_capture_kms_capture(gsr_capture *cap, AVFrame *frame, gsr_color_c
     output_size = scale_keep_aspect_ratio(self->capture_size, output_size);
 
     const float texture_rotation = monitor_rotation_to_radians(self->monitor_rotation);
-    const vec2i target_pos = { max_int(0, frame->width / 2 - output_size.x / 2), max_int(0, frame->height / 2 - output_size.y / 2) };
+    const vec2i target_pos = { max_int(0, capture_metadata->width / 2 - output_size.x / 2), max_int(0, capture_metadata->height / 2 - output_size.y / 2) };
     gsr_capture_kms_update_capture_size_change(self, color_conversion, target_pos, drm_fd);
 
     vec2i capture_pos = self->capture_pos;
@@ -675,7 +667,7 @@ static int gsr_capture_kms_capture(gsr_capture *cap, AVFrame *frame, gsr_color_c
     self->params.egl->glFinish();
 
     /* Fast opengl free path */
-    if(!self->fast_path_failed && self->monitor_rotation == GSR_MONITOR_ROT_0 && video_codec_context_is_vaapi(self->video_codec_context) && self->params.egl->gpu_info.vendor == GSR_GPU_VENDOR_AMD) {
+    if(!self->fast_path_failed && self->monitor_rotation == GSR_MONITOR_ROT_0 && video_codec_context_is_vaapi(capture_metadata->video_codec_context) && self->params.egl->gpu_info.vendor == GSR_GPU_VENDOR_AMD) {
         int fds[4];
         uint32_t offsets[4];
         uint32_t pitches[4];
@@ -686,7 +678,7 @@ static int gsr_capture_kms_capture(gsr_capture *cap, AVFrame *frame, gsr_color_c
             pitches[i] = drm_fd->dma_buf[i].pitch;
             modifiers[i] = drm_fd->modifier;
         }
-        if(!vaapi_copy_drm_planes_to_video_surface(self->video_codec_context, frame, (vec2i){capture_pos.x, capture_pos.y}, self->capture_size, target_pos, output_size, drm_fd->pixel_format, (vec2i){drm_fd->width, drm_fd->height}, fds, offsets, pitches, modifiers, drm_fd->num_dma_bufs)) {
+        if(!vaapi_copy_drm_planes_to_video_surface(capture_metadata->video_codec_context, capture_metadata->frame, (vec2i){capture_pos.x, capture_pos.y}, self->capture_size, target_pos, output_size, drm_fd->pixel_format, (vec2i){drm_fd->width, drm_fd->height}, fds, offsets, pitches, modifiers, drm_fd->num_dma_bufs)) {
             fprintf(stderr, "gsr error: gsr_capture_kms_capture: vaapi_copy_drm_planes_to_video_surface failed, falling back to opengl copy. Please report this as an issue at https://github.com/dec05eba/gpu-screen-recorder-issues\n");
             self->fast_path_failed = true;
         }
@@ -777,8 +769,7 @@ static bool gsr_capture_kms_set_hdr_metadata(gsr_capture *cap, AVMasteringDispla
 //     self->damaged = false;
 // }
 
-static void gsr_capture_kms_destroy(gsr_capture *cap, AVCodecContext *video_codec_context) {
-    (void)video_codec_context;
+static void gsr_capture_kms_destroy(gsr_capture *cap) {
     gsr_capture_kms *self = cap->priv;
     if(cap->priv) {
         gsr_capture_kms_stop(self);
diff --git a/src/capture/nvfbc.c b/src/capture/nvfbc.c
index 676d269..af79e0d 100644
--- a/src/capture/nvfbc.c
+++ b/src/capture/nvfbc.c
@@ -133,31 +133,6 @@ static bool gsr_capture_nvfbc_load_library(gsr_capture *cap) {
     return true;
 }
 
-/* TODO: check for glx swap control extension string (GLX_EXT_swap_control, etc) */
-static void set_vertical_sync_enabled(gsr_egl *egl, int enabled) {
-    int result = 0;
-
-    if(egl->glXSwapIntervalEXT) {
-        assert(gsr_window_get_display_server(egl->window) == GSR_DISPLAY_SERVER_X11);
-        Display *display = gsr_window_get_display(egl->window);
-        const Window window = (Window)gsr_window_get_window(egl->window);
-        egl->glXSwapIntervalEXT(display, window, enabled ? 1 : 0);
-    } else if(egl->glXSwapIntervalMESA) {
-        result = egl->glXSwapIntervalMESA(enabled ? 1 : 0);
-    } else if(egl->glXSwapIntervalSGI) {
-        result = egl->glXSwapIntervalSGI(enabled ? 1 : 0);
-    } else {
-        static int warned = 0;
-        if (!warned) {
-            warned = 1;
-            fprintf(stderr, "gsr warning: setting vertical sync not supported\n");
-        }
-    }
-
-    if(result != 0)
-        fprintf(stderr, "gsr warning: setting vertical sync failed\n");
-}
-
 static void gsr_capture_nvfbc_destroy_session(gsr_capture_nvfbc *self) {
     if(self->fbc_handle_created && self->capture_session_created) {
         NVFBC_DESTROY_CAPTURE_SESSION_PARAMS destroy_capture_params;
@@ -311,7 +286,7 @@ static void gsr_capture_nvfbc_stop(gsr_capture_nvfbc *self) {
     }
 }
 
-static int gsr_capture_nvfbc_start(gsr_capture *cap, AVCodecContext *video_codec_context, AVFrame *frame) {
+static int gsr_capture_nvfbc_start(gsr_capture *cap, gsr_capture_metadata *capture_metadata) {
     gsr_capture_nvfbc *self = cap->priv;
 
     if(!gsr_capture_nvfbc_load_library(cap))
@@ -357,27 +332,21 @@ static int gsr_capture_nvfbc_start(gsr_capture *cap, AVCodecContext *video_codec
     }
 
     if(self->capture_region) {
-        video_codec_context->width = FFALIGN(self->width, 2);
-        video_codec_context->height = FFALIGN(self->height, 2);
+        capture_metadata->width = FFALIGN(self->width, 2);
+        capture_metadata->height = FFALIGN(self->height, 2);
     } else {
-        video_codec_context->width = FFALIGN(self->tracking_width, 2);
-        video_codec_context->height = FFALIGN(self->tracking_height, 2);
+        capture_metadata->width = FFALIGN(self->tracking_width, 2);
+        capture_metadata->height = FFALIGN(self->tracking_height, 2);
     }
 
     if(self->params.output_resolution.x == 0 && self->params.output_resolution.y == 0) {
-        self->params.output_resolution = (vec2i){video_codec_context->width, video_codec_context->height};
+        self->params.output_resolution = (vec2i){capture_metadata->width, capture_metadata->height};
     } else {
-        self->params.output_resolution = scale_keep_aspect_ratio((vec2i){video_codec_context->width, video_codec_context->height}, self->params.output_resolution);
-        video_codec_context->width = FFALIGN(self->params.output_resolution.x, 2);
-        video_codec_context->height = FFALIGN(self->params.output_resolution.y, 2);
+        self->params.output_resolution = scale_keep_aspect_ratio((vec2i){capture_metadata->width, capture_metadata->height}, self->params.output_resolution);
+        capture_metadata->width = FFALIGN(self->params.output_resolution.x, 2);
+        capture_metadata->height = FFALIGN(self->params.output_resolution.y, 2);
     }
 
-    frame->width = video_codec_context->width;
-    frame->height = video_codec_context->height;
-
-    /* Disable vsync */
-    set_vertical_sync_enabled(self->params.egl, 0);
-
     return 0;
 
     error_cleanup:
@@ -385,7 +354,7 @@ static int gsr_capture_nvfbc_start(gsr_capture *cap, AVCodecContext *video_codec
     return -1;
 }
 
-static int gsr_capture_nvfbc_capture(gsr_capture *cap, AVFrame *frame, gsr_color_conversion *color_conversion) {
+static int gsr_capture_nvfbc_capture(gsr_capture *cap, gsr_capture_metadata *capture_metadata, gsr_color_conversion *color_conversion) {
     gsr_capture_nvfbc *self = cap->priv;
 
     const double nvfbc_recreate_retry_time_seconds = 1.0;
@@ -416,7 +385,7 @@ static int gsr_capture_nvfbc_capture(gsr_capture *cap, AVFrame *frame, gsr_color
     vec2i output_size = is_scaled ? self->params.output_resolution : frame_size;
     output_size = scale_keep_aspect_ratio(frame_size, output_size);
 
-    const vec2i target_pos = { max_int(0, frame->width / 2 - output_size.x / 2), max_int(0, frame->height / 2 - output_size.y / 2) };
+    const vec2i target_pos = { max_int(0, capture_metadata->width / 2 - output_size.x / 2), max_int(0, capture_metadata->height / 2 - output_size.y / 2) };
 
     NVFBC_FRAME_GRAB_INFO frame_info;
     memset(&frame_info, 0, sizeof(frame_info));
@@ -450,8 +419,7 @@ static int gsr_capture_nvfbc_capture(gsr_capture *cap, AVFrame *frame, gsr_color
     return 0;
 }
 
-static void gsr_capture_nvfbc_destroy(gsr_capture *cap, AVCodecContext *video_codec_context) {
-    (void)video_codec_context;
+static void gsr_capture_nvfbc_destroy(gsr_capture *cap) {
     gsr_capture_nvfbc *self = cap->priv;
     gsr_capture_nvfbc_stop(self);
     free(cap->priv);
diff --git a/src/capture/portal.c b/src/capture/portal.c
index 27486fd..cfbfbcd 100644
--- a/src/capture/portal.c
+++ b/src/capture/portal.c
@@ -25,7 +25,6 @@ typedef struct {
     gsr_pipewire_video_dmabuf_data dmabuf_data[GSR_PIPEWIRE_VIDEO_DMABUF_MAX_PLANES];
     int num_dmabuf_data;
 
-    AVCodecContext *video_codec_context;
     bool fast_path_failed;
     bool mesa_supports_compute_only_vaapi_copy;
 } gsr_capture_portal;
@@ -257,7 +256,7 @@ static bool gsr_capture_portal_get_frame_dimensions(gsr_capture_portal *self) {
     return false;
 }
 
-static int gsr_capture_portal_start(gsr_capture *cap, AVCodecContext *video_codec_context, AVFrame *frame) {
+static int gsr_capture_portal_start(gsr_capture *cap, gsr_capture_metadata *capture_metadata) {
     gsr_capture_portal *self = cap->priv;
 
     gsr_capture_portal_create_input_textures(self);
@@ -286,7 +285,7 @@ static int gsr_capture_portal_start(gsr_capture *cap, AVCodecContext *video_code
     fprintf(stderr, "gsr info: gsr_capture_portal_start: setting up pipewire\n");
     /* TODO: support hdr when pipewire supports it */
     /* gsr_pipewire closes the pipewire fd, even on failure */
-    if(!gsr_pipewire_video_init(&self->pipewire, pipewire_fd, pipewire_node, video_codec_context->framerate.num, self->params.record_cursor, self->params.egl)) {
+    if(!gsr_pipewire_video_init(&self->pipewire, pipewire_fd, pipewire_node, capture_metadata->fps, self->params.record_cursor, self->params.egl)) {
         fprintf(stderr, "gsr error: gsr_capture_portal_start: failed to setup pipewire with fd: %d, node: %" PRIu32 "\n", pipewire_fd, pipewire_node);
         gsr_capture_portal_stop(self);
         return -1;
@@ -298,17 +297,14 @@ static int gsr_capture_portal_start(gsr_capture *cap, AVCodecContext *video_code
         return -1;
     }
 
-    /* Disable vsync */
-    self->params.egl->eglSwapInterval(self->params.egl->egl_display, 0);
-
     if(self->params.output_resolution.x == 0 && self->params.output_resolution.y == 0) {
         self->params.output_resolution = self->capture_size;
-        video_codec_context->width = FFALIGN(self->capture_size.x, 2);
-        video_codec_context->height = FFALIGN(self->capture_size.y, 2);
+        capture_metadata->width = FFALIGN(self->capture_size.x, 2);
+        capture_metadata->height = FFALIGN(self->capture_size.y, 2);
     } else {
         self->params.output_resolution = scale_keep_aspect_ratio(self->capture_size, self->params.output_resolution);
-        video_codec_context->width = FFALIGN(self->params.output_resolution.x, 2);
-        video_codec_context->height = FFALIGN(self->params.output_resolution.y, 2);
+        capture_metadata->width = FFALIGN(self->params.output_resolution.x, 2);
+        capture_metadata->height = FFALIGN(self->params.output_resolution.y, 2);
     }
 
     self->fast_path_failed = self->params.egl->gpu_info.vendor == GSR_GPU_VENDOR_AMD && !gl_driver_version_greater_than(&self->params.egl->gpu_info, 24, 0, 9);
@@ -317,10 +313,6 @@ static int gsr_capture_portal_start(gsr_capture *cap, AVCodecContext *video_code
 
     self->mesa_supports_compute_only_vaapi_copy = self->params.egl->gpu_info.vendor == GSR_GPU_VENDOR_AMD && gl_driver_version_greater_than(&self->params.egl->gpu_info, 24, 3, 6);
 
-    frame->width = video_codec_context->width;
-    frame->height = video_codec_context->height;
-
-    self->video_codec_context = video_codec_context;
     return 0;
 }
 
@@ -338,8 +330,7 @@ static void gsr_capture_portal_fail_fast_path_if_not_fast(gsr_capture_portal *se
     }
 }
 
-static int gsr_capture_portal_capture(gsr_capture *cap, AVFrame *frame, gsr_color_conversion *color_conversion) {
-    (void)frame;
+static int gsr_capture_portal_capture(gsr_capture *cap, gsr_capture_metadata *capture_metadata, gsr_color_conversion *color_conversion) {
     (void)color_conversion;
     gsr_capture_portal *self = cap->priv;
 
@@ -365,7 +356,7 @@ static int gsr_capture_portal_capture(gsr_capture *cap, AVFrame *frame, gsr_colo
     vec2i output_size = is_scaled ? self->params.output_resolution : self->capture_size;
     output_size = scale_keep_aspect_ratio(self->capture_size, output_size);
     
-    const vec2i target_pos = { max_int(0, frame->width / 2 - output_size.x / 2), max_int(0, frame->height / 2 - output_size.y / 2) };
+    const vec2i target_pos = { max_int(0, capture_metadata->width / 2 - output_size.x / 2), max_int(0, capture_metadata->height / 2 - output_size.y / 2) };
 
     self->params.egl->glFlush();
     self->params.egl->glFinish();
@@ -373,7 +364,7 @@ static int gsr_capture_portal_capture(gsr_capture *cap, AVFrame *frame, gsr_colo
     // TODO: Handle region crop
 
     /* Fast opengl free path */
-    if(!self->fast_path_failed && video_codec_context_is_vaapi(self->video_codec_context) && self->params.egl->gpu_info.vendor == GSR_GPU_VENDOR_AMD) {
+    if(!self->fast_path_failed && video_codec_context_is_vaapi(capture_metadata->video_codec_context) && self->params.egl->gpu_info.vendor == GSR_GPU_VENDOR_AMD) {
         int fds[4];
         uint32_t offsets[4];
         uint32_t pitches[4];
@@ -384,7 +375,7 @@ static int gsr_capture_portal_capture(gsr_capture *cap, AVFrame *frame, gsr_colo
             pitches[i] = self->dmabuf_data[i].stride;
             modifiers[i] = pipewire_modifiers;
         }
-        if(!vaapi_copy_drm_planes_to_video_surface(self->video_codec_context, frame, (vec2i){region.x, region.y}, self->capture_size, target_pos, output_size, pipewire_fourcc, self->capture_size, fds, offsets, pitches, modifiers, self->num_dmabuf_data)) {
+        if(!vaapi_copy_drm_planes_to_video_surface(capture_metadata->video_codec_context, capture_metadata->frame, (vec2i){region.x, region.y}, self->capture_size, target_pos, output_size, pipewire_fourcc, self->capture_size, fds, offsets, pitches, modifiers, self->num_dmabuf_data)) {
             fprintf(stderr, "gsr error: gsr_capture_portal_capture: vaapi_copy_drm_planes_to_video_surface failed, falling back to opengl copy. Please report this as an issue at https://github.com/dec05eba/gpu-screen-recorder-issues\n");
             self->fast_path_failed = true;
         }
@@ -442,8 +433,7 @@ static void gsr_capture_portal_clear_damage(gsr_capture *cap) {
     gsr_pipewire_video_clear_damage(&self->pipewire);
 }
 
-static void gsr_capture_portal_destroy(gsr_capture *cap, AVCodecContext *video_codec_context) {
-    (void)video_codec_context;
+static void gsr_capture_portal_destroy(gsr_capture *cap) {
     gsr_capture_portal *self = cap->priv;
     if(cap->priv) {
         gsr_capture_portal_stop(self);
diff --git a/src/capture/xcomposite.c b/src/capture/xcomposite.c
index 5cef71d..94e691b 100644
--- a/src/capture/xcomposite.c
+++ b/src/capture/xcomposite.c
@@ -31,7 +31,6 @@ typedef struct {
     double window_resize_timer;
     
     WindowTexture window_texture;
-    AVCodecContext *video_codec_context;
 
     Atom net_active_window_atom;
 
@@ -64,7 +63,7 @@ static Window get_focused_window(Display *display, Atom net_active_window_atom)
     return None;
 }
 
-static int gsr_capture_xcomposite_start(gsr_capture *cap, AVCodecContext *video_codec_context, AVFrame *frame) {
+static int gsr_capture_xcomposite_start(gsr_capture *cap, gsr_capture_metadata *capture_metadata) {
     gsr_capture_xcomposite *self = cap->priv;
 
     if(self->params.follow_focused) {
@@ -95,8 +94,6 @@ static int gsr_capture_xcomposite_start(gsr_capture *cap, AVCodecContext *video_
     // TODO: Get select and add these on top of it and then restore at the end. Also do the same in other xcomposite
     XSelectInput(self->display, self->window, StructureNotifyMask | ExposureMask);
 
-    /* Disable vsync */
-    self->params.egl->eglSwapInterval(self->params.egl->egl_display, 0);
     if(window_texture_init(&self->window_texture, self->display, self->window, self->params.egl) != 0 && !self->params.follow_focused) {
         fprintf(stderr, "gsr error: gsr_capture_xcomposite_start: failed to get window texture for window %ld\n", (long)self->window);
         return -1;
@@ -117,21 +114,17 @@ static int gsr_capture_xcomposite_start(gsr_capture *cap, AVCodecContext *video_
 
     if(self->params.output_resolution.x == 0 && self->params.output_resolution.y == 0) {
         self->params.output_resolution = self->texture_size;
-        video_codec_context->width = FFALIGN(self->texture_size.x, 2);
-        video_codec_context->height = FFALIGN(self->texture_size.y, 2);
+        capture_metadata->width = FFALIGN(self->texture_size.x, 2);
+        capture_metadata->height = FFALIGN(self->texture_size.y, 2);
     } else {
-        video_codec_context->width = FFALIGN(self->params.output_resolution.x, 2);
-        video_codec_context->height = FFALIGN(self->params.output_resolution.y, 2);
+        capture_metadata->width = FFALIGN(self->params.output_resolution.x, 2);
+        capture_metadata->height = FFALIGN(self->params.output_resolution.y, 2);
     }
 
     self->fast_path_failed = self->params.egl->gpu_info.vendor == GSR_GPU_VENDOR_AMD && !gl_driver_version_greater_than(&self->params.egl->gpu_info, 24, 0, 9);
     if(self->fast_path_failed)
         fprintf(stderr, "gsr warning: gsr_capture_kms_start: your amd driver (mesa) version is known to be buggy (<= version 24.0.9), falling back to opengl copy\n");
 
-    frame->width = video_codec_context->width;
-    frame->height = video_codec_context->height;
-
-    self->video_codec_context = video_codec_context;
     self->window_resize_timer = clock_get_monotonic_seconds();
     return 0;
 }
@@ -255,9 +248,8 @@ static bool gsr_capture_xcomposite_should_stop(gsr_capture *cap, bool *err) {
     return false;
 }
 
-static int gsr_capture_xcomposite_capture(gsr_capture *cap, AVFrame *frame, gsr_color_conversion *color_conversion) {
+static int gsr_capture_xcomposite_capture(gsr_capture *cap, gsr_capture_metadata *capture_metdata, gsr_color_conversion *color_conversion) {
     gsr_capture_xcomposite *self = cap->priv;
-    (void)frame;
 
     if(self->clear_background) {
         self->clear_background = false;
@@ -268,14 +260,14 @@ static int gsr_capture_xcomposite_capture(gsr_capture *cap, AVFrame *frame, gsr_
     vec2i output_size = is_scaled ? self->params.output_resolution : self->texture_size;
     output_size = scale_keep_aspect_ratio(self->texture_size, output_size);
 
-    const vec2i target_pos = { max_int(0, frame->width / 2 - output_size.x / 2), max_int(0, frame->height / 2 - output_size.y / 2) };
+    const vec2i target_pos = { max_int(0, capture_metdata->width / 2 - output_size.x / 2), max_int(0, capture_metdata->height / 2 - output_size.y / 2) };
 
     self->params.egl->glFlush();
     self->params.egl->glFinish();
 
     /* Fast opengl free path */
-    if(!self->fast_path_failed && video_codec_context_is_vaapi(self->video_codec_context) && self->params.egl->gpu_info.vendor == GSR_GPU_VENDOR_AMD) {
-        if(!vaapi_copy_egl_image_to_video_surface(self->params.egl, self->window_texture.image, (vec2i){0, 0}, self->texture_size, target_pos, output_size, self->video_codec_context, frame)) {
+    if(!self->fast_path_failed && video_codec_context_is_vaapi(capture_metdata->video_codec_context) && self->params.egl->gpu_info.vendor == GSR_GPU_VENDOR_AMD) {
+        if(!vaapi_copy_egl_image_to_video_surface(self->params.egl, self->window_texture.image, (vec2i){0, 0}, self->texture_size, target_pos, output_size, capture_metdata->video_codec_context, capture_metdata->frame)) {
             fprintf(stderr, "gsr error: gsr_capture_xcomposite_capture: vaapi_copy_egl_image_to_video_surface failed, falling back to opengl copy. Please report this as an issue at https://github.com/dec05eba/gpu-screen-recorder-issues\n");
             self->fast_path_failed = true;
         }
@@ -325,8 +317,7 @@ static uint64_t gsr_capture_xcomposite_get_window_id(gsr_capture *cap) {
     return self->window;
 }
 
-static void gsr_capture_xcomposite_destroy(gsr_capture *cap, AVCodecContext *video_codec_context) {
-    (void)video_codec_context;
+static void gsr_capture_xcomposite_destroy(gsr_capture *cap) {
     if(cap->priv) {
         gsr_capture_xcomposite_stop(cap->priv);
         free(cap->priv);
diff --git a/src/egl.c b/src/egl.c
index 682810d..be5a44d 100644
--- a/src/egl.c
+++ b/src/egl.c
@@ -355,6 +355,44 @@ static void debug_callback(unsigned int source, unsigned int type, unsigned int
         fprintf(stderr, "gsr info: gl callback: %s type = 0x%x, severity = 0x%x, message = %s\n", type == GL_DEBUG_TYPE_ERROR ? "** GL ERROR **" : "", type, severity, message);
 }
 
+/* TODO: check for glx swap control extension string (GLX_EXT_swap_control, etc) */
+static void set_vertical_sync_enabled(gsr_egl *egl, int enabled) {
+    int result = 0;
+
+    if(egl->glXSwapIntervalEXT) {
+        assert(gsr_window_get_display_server(egl->window) == GSR_DISPLAY_SERVER_X11);
+        Display *display = gsr_window_get_display(egl->window);
+        const Window window = (Window)gsr_window_get_window(egl->window);
+        egl->glXSwapIntervalEXT(display, window, enabled ? 1 : 0);
+    } else if(egl->glXSwapIntervalMESA) {
+        result = egl->glXSwapIntervalMESA(enabled ? 1 : 0);
+    } else if(egl->glXSwapIntervalSGI) {
+        result = egl->glXSwapIntervalSGI(enabled ? 1 : 0);
+    } else {
+        static int warned = 0;
+        if (!warned) {
+            warned = 1;
+            fprintf(stderr, "gsr warning: setting vertical sync not supported\n");
+        }
+    }
+
+    if(result != 0)
+        fprintf(stderr, "gsr warning: setting vertical sync failed\n");
+}
+
+static void gsr_egl_disable_vsync(gsr_egl *self) {
+    switch(self->context_type) {
+        case GSR_GL_CONTEXT_TYPE_EGL: {
+            self->eglSwapInterval(self->egl_display, 0);
+            break;
+        }
+        case GSR_GL_CONTEXT_TYPE_GLX: {
+            set_vertical_sync_enabled(self, 0);
+            break;
+        }
+    }
+}
+
 bool gsr_egl_load(gsr_egl *self, gsr_window *window, bool is_monitor_capture, bool enable_debug) {
     memset(self, 0, sizeof(gsr_egl));
     self->context_type = GSR_GL_CONTEXT_TYPE_EGL;
@@ -416,6 +454,7 @@ bool gsr_egl_load(gsr_egl *self, gsr_window *window, bool is_monitor_capture, bo
         self->glDebugMessageCallback(debug_callback, NULL);
     }
 
+    gsr_egl_disable_vsync(self);
     return true;
 
     fail:
diff --git a/src/encoder/video/image.c b/src/encoder/video/image.c
deleted file mode 100644
index 5f48b7b..0000000
--- a/src/encoder/video/image.c
+++ /dev/null
@@ -1,130 +0,0 @@
-#include "../../../include/encoder/video/image.h"
-#include "../../../include/egl.h"
-
-#include <libavcodec/avcodec.h>
-#include <libavutil/frame.h>
-
-#include <stdlib.h>
-
-#define LINESIZE_ALIGNMENT 4
-
-typedef struct {
-    gsr_video_encoder_image_params params;
-
-    unsigned int target_texture;
-} gsr_video_encoder_image;
-
-static unsigned int gl_create_texture(gsr_egl *egl, int width, int height, int internal_format, unsigned int format) {
-    unsigned int texture_id = 0;
-    egl->glGenTextures(1, &texture_id);
-    egl->glBindTexture(GL_TEXTURE_2D, texture_id);
-    egl->glTexImage2D(GL_TEXTURE_2D, 0, internal_format, width, height, 0, format, GL_UNSIGNED_BYTE, NULL);
-
-    egl->glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_S, GL_CLAMP_TO_EDGE);
-    egl->glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_T, GL_CLAMP_TO_EDGE);
-    egl->glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_NEAREST);
-    egl->glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_NEAREST);
-
-    egl->glBindTexture(GL_TEXTURE_2D, 0);
-    return texture_id;
-}
-
-static bool gsr_video_encoder_image_setup_textures(gsr_video_encoder_image *self, AVCodecContext *video_codec_context, AVFrame *frame) {
-    int res = av_frame_get_buffer(frame, LINESIZE_ALIGNMENT);
-    if(res < 0) {
-        fprintf(stderr, "gsr error: gsr_video_encoder_image_setup_textures: av_frame_get_buffer failed: %d\n", res);
-        return false;
-    }
-
-    res = av_frame_make_writable(frame);
-    if(res < 0) {
-        fprintf(stderr, "gsr error: gsr_video_encoder_image_setup_textures: av_frame_make_writable failed: %d\n", res);
-        return false;
-    }
-
-    self->target_texture = gl_create_texture(self->params.egl, video_codec_context->width, video_codec_context->height, self->params.color_depth == GSR_COLOR_DEPTH_8_BITS ? GL_RGB8 : GL_RGB16, GL_RGB);
-    if(self->target_texture == 0) {
-        fprintf(stderr, "gsr error: gsr_capture_kms_setup_cuda_textures: failed to create opengl texture\n");
-        return false;
-    }
-
-    return true;
-}
-
-static void gsr_video_encoder_image_stop(gsr_video_encoder_image *self, AVCodecContext *video_codec_context);
-
-static bool gsr_video_encoder_image_start(gsr_video_encoder *encoder, AVCodecContext *video_codec_context, AVFrame *frame) {
-    gsr_video_encoder_image *self = encoder->priv;
-
-    video_codec_context->width = FFALIGN(video_codec_context->width, LINESIZE_ALIGNMENT);
-    video_codec_context->height = FFALIGN(video_codec_context->height, 2);
-
-    frame->width = video_codec_context->width;
-    frame->height = video_codec_context->height;
-
-    if(!gsr_video_encoder_image_setup_textures(self, video_codec_context, frame)) {
-        gsr_video_encoder_image_stop(self, video_codec_context);
-        return false;
-    }
-
-    return true;
-}
-
-void gsr_video_encoder_image_stop(gsr_video_encoder_image *self, AVCodecContext *video_codec_context) {
-    (void)video_codec_context;
-    self->params.egl->glDeleteTextures(1, &self->target_texture);
-    self->target_texture = 0;
-}
-
-static void gsr_video_encoder_image_copy_textures_to_frame(gsr_video_encoder *encoder, AVFrame *frame, gsr_color_conversion *color_conversion) {
-    gsr_video_encoder_image *self = encoder->priv;
-    // TODO: hdr support
-    self->params.egl->glBindTexture(GL_TEXTURE_2D, self->target_texture);
-    // We could use glGetTexSubImage and then we wouldn't have to use a specific linesize (LINESIZE_ALIGNMENT) that adds padding,
-    // but glGetTexSubImage is only available starting from opengl 4.5.
-    self->params.egl->glGetTexImage(GL_TEXTURE_2D, 0, GL_RGB, GL_UNSIGNED_BYTE, frame->data[0]);
-    self->params.egl->glBindTexture(GL_TEXTURE_2D, 0);
-    // cap_kms->kms.base.egl->eglSwapBuffers(cap_kms->kms.base.egl->egl_display, cap_kms->kms.base.egl->egl_surface);
-
-    self->params.egl->glFlush();
-    self->params.egl->glFinish();
-}
-
-static void gsr_video_encoder_image_get_textures(gsr_video_encoder *encoder, unsigned int *textures, int *num_textures, gsr_destination_color *destination_color) {
-    gsr_video_encoder_image *self = encoder->priv;
-    textures[0] = self->target_texture;
-    *num_textures = 1;
-    // TODO: 10-bit support
-    //*destination_color = self->params.color_depth == GSR_COLOR_DEPTH_10_BITS ? GSR_DESTINATION_COLOR_P010 : GSR_DESTINATION_COLOR_NV12;
-    *destination_color = GSR_DESTINATION_COLOR_RGB8;
-}
-
-static void gsr_video_encoder_image_destroy(gsr_video_encoder *encoder, AVCodecContext *video_codec_context) {
-    gsr_video_encoder_image_stop(encoder->priv, video_codec_context);
-    free(encoder->priv);
-    free(encoder);
-}
-
-gsr_video_encoder* gsr_video_encoder_image_create(const gsr_video_encoder_image_params *params) {
-    gsr_video_encoder *encoder = calloc(1, sizeof(gsr_video_encoder));
-    if(!encoder)
-        return NULL;
-
-    gsr_video_encoder_image *encoder_image = calloc(1, sizeof(gsr_video_encoder_image));
-    if(!encoder_image) {
-        free(encoder);
-        return NULL;
-    }
-
-    encoder_image->params = *params;
-
-    *encoder = (gsr_video_encoder) {
-        .start = gsr_video_encoder_image_start,
-        .copy_textures_to_frame = gsr_video_encoder_image_copy_textures_to_frame,
-        .get_textures = gsr_video_encoder_image_get_textures,
-        .destroy = gsr_video_encoder_image_destroy,
-        .priv = encoder_image
-    };
-
-    return encoder;
-}
diff --git a/src/image_writer.c b/src/image_writer.c
new file mode 100644
index 0000000..d01a66c
--- /dev/null
+++ b/src/image_writer.c
@@ -0,0 +1,85 @@
+#include "../include/image_writer.h"
+#include "../include/egl.h"
+
+#define STB_IMAGE_WRITE_IMPLEMENTATION
+#include "../external/stb_image_write.h"
+
+#include <stdlib.h>
+#include <stdint.h>
+#include <stdio.h>
+#include <assert.h>
+
+static unsigned int gl_create_texture(gsr_egl *egl, int width, int height, int internal_format, unsigned int format) {
+    unsigned int texture_id = 0;
+    egl->glGenTextures(1, &texture_id);
+    egl->glBindTexture(GL_TEXTURE_2D, texture_id);
+    egl->glTexImage2D(GL_TEXTURE_2D, 0, internal_format, width, height, 0, format, GL_UNSIGNED_BYTE, NULL);
+
+    egl->glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_S, GL_CLAMP_TO_EDGE);
+    egl->glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_T, GL_CLAMP_TO_EDGE);
+    egl->glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_NEAREST);
+    egl->glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_NEAREST);
+
+    egl->glBindTexture(GL_TEXTURE_2D, 0);
+    return texture_id;
+}
+
+/* TODO: Support hdr/10-bit */
+bool gsr_image_writer_init(gsr_image_writer *self, gsr_image_writer_source source, gsr_egl *egl, int width, int height) {
+    assert(source == GSR_IMAGE_WRITER_SOURCE_OPENGL);
+    self->source = source;
+    self->egl = egl;
+    self->width = width;
+    self->height = height;
+    self->texture = gl_create_texture(self->egl, self->width, self->height, GL_RGB8, GL_RGB); /* TODO: use GL_RGB16 instead of GL_RGB8 for hdr/10-bit */
+    if(self->texture == 0) {
+        fprintf(stderr, "gsr error: gsr_image_writer_init: failed to create texture\n");
+        return false;
+    }
+    return true;
+}
+
+void gsr_image_writer_deinit(gsr_image_writer *self) {
+    if(self->texture) {
+        self->egl->glDeleteTextures(1, &self->texture);
+        self->texture = 0;
+    }
+}
+
+bool gsr_image_writer_write_to_file(gsr_image_writer *self, const char *filepath, gsr_image_format image_format, int quality) {
+    if(quality < 1)
+        quality = 1;
+    else if(quality > 100)
+        quality = 100;
+
+    uint8_t *frame_data = malloc(self->width * self->height * 3);
+    if(!frame_data) {
+        fprintf(stderr, "gsr error: gsr_image_writer_write_to_file: failed to allocate memory for image frame\n");
+        return false;
+    }
+
+    // TODO: hdr support
+    self->egl->glBindTexture(GL_TEXTURE_2D, self->texture);
+    // We could use glGetTexSubImage, but it's only available starting from opengl 4.5
+    self->egl->glGetTexImage(GL_TEXTURE_2D, 0, GL_RGB, GL_UNSIGNED_BYTE, frame_data);
+    self->egl->glBindTexture(GL_TEXTURE_2D, 0);
+
+    self->egl->glFlush();
+    self->egl->glFinish();
+    
+    bool success = false;
+    switch(image_format) {
+        case GSR_IMAGE_FORMAT_JPEG:
+            success = stbi_write_jpg(filepath, self->width, self->height, 3, frame_data, quality);
+            break;
+        case GSR_IMAGE_FORMAT_PNG:
+            success = stbi_write_png(filepath, self->width, self->height, 3, frame_data, 0);
+            break;
+    }
+
+    if(!success)
+        fprintf(stderr, "gsr error: gsr_image_writer_write_to_file: failed to write image data to output file %s\n", filepath);
+
+    free(frame_data);
+    return success;
+}
diff --git a/src/main.cpp b/src/main.cpp
index 2eb348d..10dc5a6 100644
--- a/src/main.cpp
+++ b/src/main.cpp
@@ -13,7 +13,6 @@ extern "C" {
 #include "../include/encoder/video/vaapi.h"
 #include "../include/encoder/video/vulkan.h"
 #include "../include/encoder/video/software.h"
-#include "../include/encoder/video/image.h"
 #include "../include/codec_query/nvenc.h"
 #include "../include/codec_query/vaapi.h"
 #include "../include/codec_query/vulkan.h"
@@ -23,6 +22,7 @@ extern "C" {
 #include "../include/utils.h"
 #include "../include/damage.h"
 #include "../include/color_conversion.h"
+#include "../include/image_writer.h"
 }
 
 #include <assert.h>
@@ -113,9 +113,7 @@ enum class VideoCodec {
     VP8,
     VP9,
     H264_VULKAN,
-    HEVC_VULKAN,
-    JPEG,
-    PNG
+    HEVC_VULKAN
 };
 
 enum class AudioCodec {
@@ -219,16 +217,6 @@ static bool video_codec_is_vulkan(VideoCodec video_codec) {
     }
 }
 
-static bool video_codec_is_image(VideoCodec video_codec) {
-    switch(video_codec) {
-        case VideoCodec::JPEG:
-        case VideoCodec::PNG:
-            return true;
-        default:
-            return false;
-    }
-}
-
 struct PacketData {
     PacketData() {}
     PacketData(const PacketData&) = delete;
@@ -593,22 +581,7 @@ static AVCodecContext *create_video_codec_context(AVPixelFormat pix_fmt,
     if (codec_context->codec_id == AV_CODEC_ID_MPEG1VIDEO)
         codec_context->mb_decision = 2;
 
-    if(video_codec_is_image(video_codec)) {
-        switch(video_quality) {
-            case VideoQuality::MEDIUM:
-                codec_context->compression_level = 8;
-                break;
-            case VideoQuality::HIGH:
-                codec_context->compression_level = 6;
-                break;
-            case VideoQuality::VERY_HIGH:
-                codec_context->compression_level = 4;
-                break;
-            case VideoQuality::ULTRA:
-                codec_context->compression_level = 2;
-                break;
-        }
-    } else if(!use_software_video_encoder && vendor != GSR_GPU_VENDOR_NVIDIA && bitrate_mode != BitrateMode::CBR) {
+    if(!use_software_video_encoder && vendor != GSR_GPU_VENDOR_NVIDIA && bitrate_mode != BitrateMode::CBR) {
         // 8 bit / 10 bit = 80%, and increase it even more
         const float quality_multiply = hdr ? (8.0f/10.0f * 0.7f) : 1.0f;
         if(codec_context->codec_id == AV_CODEC_ID_AV1 || codec_context->codec_id == AV_CODEC_ID_H264 || codec_context->codec_id == AV_CODEC_ID_HEVC) {
@@ -743,15 +716,6 @@ static AVFrame* create_audio_frame(AVCodecContext *audio_codec_context) {
     return frame;
 }
 
-static void open_video_image(AVCodecContext *codec_context) {
-    AVDictionary *options = nullptr;
-    int ret = avcodec_open2(codec_context, codec_context->codec, &options);
-    if (ret < 0) {
-        fprintf(stderr, "Error: Could not open video codec: %s\n", av_error_to_string(ret));
-        _exit(1);
-    }
-}
-
 static void dict_set_profile(AVCodecContext *codec_context, gsr_gpu_vendor vendor, gsr_color_depth color_depth, AVDictionary **options) {
     #if LIBAVCODEC_VERSION_INT < AV_VERSION_INT(61, 17, 100)
     if(codec_context->codec_id == AV_CODEC_ID_H264) {
@@ -1295,7 +1259,8 @@ static void usage_full() {
     printf("  %s -w screen -f 60 -a \"app:firefox|app:csgo\" -o \"$HOME/Videos/video.mp4\"\n", program_name);
     printf("  %s -w screen -f 60 -a \"app-inverse:firefox|app-inverse:csgo\" -o \"$HOME/Videos/video.mp4\"\n", program_name);
     printf("  %s -w screen -f 60 -a \"default-input|app-inverse:Brave\" -o \"$HOME/Videos/video.mp4\"\n", program_name);
-    printf("  %s -w screen -f 60 -o \"$HOME/Pictures/image.jpg\"\n", program_name);
+    printf("  %s -w screen -o \"$HOME/Pictures/image.jpg\"\n", program_name);
+    printf("  %s -w screen -q medium -o \"$HOME/Pictures/image.jpg\"\n", program_name);
     //fprintf(stderr, "  gpu-screen-recorder -w screen -f 60 -q ultra -pixfmt yuv444 -o video.mp4\n");
     fflush(stdout);
     _exit(1);
@@ -1895,14 +1860,6 @@ fail:
 static gsr_video_encoder* create_video_encoder(gsr_egl *egl, bool overclock, gsr_color_depth color_depth, bool use_software_video_encoder, VideoCodec video_codec) {
     gsr_video_encoder *video_encoder = nullptr;
 
-    if(video_codec_is_image(video_codec)) {
-        gsr_video_encoder_image_params params;
-        params.egl = egl;
-        params.color_depth = color_depth;
-        video_encoder = gsr_video_encoder_image_create(&params);
-        return video_encoder;
-    }
-
     if(use_software_video_encoder) {
         gsr_video_encoder_software_params params;
         params.egl = egl;
@@ -2052,10 +2009,6 @@ static const AVCodec* get_ffmpeg_video_codec(VideoCodec video_codec, gsr_gpu_ven
             return avcodec_find_encoder_by_name("h264_vulkan");
         case VideoCodec::HEVC_VULKAN:
             return avcodec_find_encoder_by_name("hevc_vulkan");
-        case VideoCodec::JPEG:
-            return avcodec_find_encoder_by_name("libopenjpeg");
-        case VideoCodec::PNG:
-            return avcodec_find_encoder_by_name("png");
     }
     return nullptr;
 }
@@ -2126,10 +2079,6 @@ static void list_supported_video_codecs(gsr_egl *egl, bool wayland) {
         puts("vp8");
     if(supported_video_codecs.vp9.supported)
         puts("vp9");
-    if(avcodec_find_encoder_by_name("libopenjpeg"))
-        puts("jpeg");
-    if(avcodec_find_encoder_by_name("png"))
-        puts("png");
     //if(supported_video_codecs_vulkan.h264.supported)
     //    puts("h264_vulkan");
     //if(supported_video_codecs_vulkan.hevc.supported)
@@ -2247,6 +2196,9 @@ static void info_command() {
     list_gpu_info(&egl);
     puts("section=video_codecs");
     list_supported_video_codecs(&egl, wayland);
+    puts("section=image_formats");
+    puts("jpeg");
+    puts("png");
     puts("section=capture_options");
     list_supported_capture_options(window, egl.card_path, list_monitors);
 
@@ -2404,7 +2356,7 @@ static void validate_monitor_get_valid(const gsr_egl *egl, std::string &window_s
     }
 }
 
-static gsr_capture* create_capture_impl(std::string &window_str, vec2i output_resolution, bool wayland, gsr_egl *egl, int fps, VideoCodec video_codec, gsr_color_range color_range,
+static gsr_capture* create_capture_impl(std::string &window_str, vec2i output_resolution, bool wayland, gsr_egl *egl, int fps, bool hdr, gsr_color_range color_range,
     bool record_cursor, bool restore_portal_session, const char *portal_session_token_filepath,
     gsr_color_depth color_depth)
 {
@@ -2478,7 +2430,7 @@ static gsr_capture* create_capture_impl(std::string &window_str, vec2i output_re
             kms_params.color_depth = color_depth;
             kms_params.color_range = color_range;
             kms_params.record_cursor = record_cursor;
-            kms_params.hdr = video_codec_is_hdr(video_codec);
+            kms_params.hdr = hdr;
             kms_params.fps = fps;
             kms_params.output_resolution = output_resolution;
             capture = gsr_capture_kms_create(&kms_params);
@@ -2516,11 +2468,107 @@ static gsr_capture* create_capture_impl(std::string &window_str, vec2i output_re
     return capture;
 }
 
+static gsr_color_range image_format_to_color_range(gsr_image_format image_format) {
+    switch(image_format) {
+        case GSR_IMAGE_FORMAT_JPEG: return GSR_COLOR_RANGE_LIMITED;
+        case GSR_IMAGE_FORMAT_PNG:  return GSR_COLOR_RANGE_FULL;
+    }
+    assert(false);
+    return GSR_COLOR_RANGE_FULL;
+}
+
+static int video_quality_to_image_quality_value(VideoQuality video_quality) {
+    switch(video_quality) {
+        case VideoQuality::MEDIUM:
+            return 60;
+        case VideoQuality::HIGH:
+            return 70;
+        case VideoQuality::VERY_HIGH:
+            return 80;
+        case VideoQuality::ULTRA:
+            return 95;
+    }
+    assert(false);
+    return 80;
+}
+
+// TODO: 10-bit and hdr.
+static void capture_image_to_file(const char *filepath, std::string &window_str, vec2i output_resolution, bool wayland, gsr_egl *egl, gsr_image_format image_format,
+    bool record_cursor, bool restore_portal_session, const char *portal_session_token_filepath, VideoQuality video_quality) {
+    const gsr_color_range color_range = image_format_to_color_range(image_format);
+    const int fps = 60;
+    gsr_capture *capture = create_capture_impl(window_str, output_resolution, wayland, egl, fps, false, color_range, record_cursor, restore_portal_session, portal_session_token_filepath, GSR_COLOR_DEPTH_8_BITS);
+
+    gsr_capture_metadata capture_metadata;
+    capture_metadata.width = 0;
+    capture_metadata.height = 0;
+    capture_metadata.fps = fps;
+    capture_metadata.video_codec_context = nullptr;
+    capture_metadata.frame = nullptr;
+
+    int capture_result = gsr_capture_start(capture, &capture_metadata);
+    if(capture_result != 0) {
+        fprintf(stderr, "gsr error: gsr_capture_start failed\n");
+        _exit(capture_result);
+    }
+
+    gsr_image_writer image_writer;
+    if(!gsr_image_writer_init(&image_writer, GSR_IMAGE_WRITER_SOURCE_OPENGL, egl, capture_metadata.width, capture_metadata.height)) {
+        fprintf(stderr, "gsr error: gsr_image_write_gl_init failed\n");
+        _exit(1);
+    }
+
+    gsr_color_conversion_params color_conversion_params;
+    memset(&color_conversion_params, 0, sizeof(color_conversion_params));
+    color_conversion_params.color_range = color_range;
+    color_conversion_params.egl = egl;
+    color_conversion_params.load_external_image_shader = gsr_capture_uses_external_image(capture);
+
+    color_conversion_params.destination_textures[0] = image_writer.texture;
+    color_conversion_params.num_destination_textures = 1;
+    color_conversion_params.destination_color = GSR_DESTINATION_COLOR_RGB8;
+
+    gsr_color_conversion color_conversion;
+    if(gsr_color_conversion_init(&color_conversion, &color_conversion_params) != 0) {
+        fprintf(stderr, "gsr error: gsr_capture_kms_setup_vaapi_textures: failed to create color conversion\n");
+        _exit(1);
+    }
+
+    gsr_color_conversion_clear(&color_conversion);
+
+    bool should_stop_error = false;
+    egl->glClear(0);
+
+    while(true) {
+        should_stop_error = false;
+        if(gsr_capture_should_stop(capture, &should_stop_error))
+            break;
+
+        // It can fail, for example when capturing portal and the target is a monitor that hasn't been updated.
+        // Desktop portal wont refresh the image until there is an update.
+        // TODO: Find out if there is a way to force update desktop portal image.
+        // This can also happen for example if the system suspends and the monitor to capture's framebuffer is gone, or if the target window disappeared.
+        if(gsr_capture_capture(capture, &capture_metadata, &color_conversion) == 0)
+            break;
+
+        usleep(30 * 1000); // 30 ms
+    }
+
+    gsr_egl_swap_buffers(egl);
+    
+    const int image_quality = video_quality_to_image_quality_value(video_quality);
+    if(!gsr_image_writer_write_to_file(&image_writer, filepath, image_format, image_quality)) {
+        fprintf(stderr, "gsr error: failed to write opengl texture to image output file %s\n", filepath);
+        _exit(1);
+    }
+
+    gsr_image_writer_deinit(&image_writer);
+    gsr_capture_destroy(capture);
+    _exit(should_stop_error ? 3 : 0);
+}
+
 static AVPixelFormat get_pixel_format(VideoCodec video_codec, gsr_gpu_vendor vendor, bool use_software_video_encoder) {
-    if(video_codec_is_image(video_codec)) {
-        // TODO: hdr
-        return AV_PIX_FMT_RGB24;
-    } else if(use_software_video_encoder) {
+    if(use_software_video_encoder) {
         return AV_PIX_FMT_NV12;
     } else {
         if(video_codec_is_vulkan(video_codec))
@@ -2744,8 +2792,6 @@ static const char* video_codec_to_string(VideoCodec video_codec) {
         case VideoCodec::VP9:         return "vp9";
         case VideoCodec::H264_VULKAN: return "h264_vulkan";
         case VideoCodec::HEVC_VULKAN: return "hevc_vulkan";
-        case VideoCodec::JPEG:        return "jpeg";
-        case VideoCodec::PNG:         return "png";
     }
     return "";
 }
@@ -2763,8 +2809,6 @@ static bool video_codec_only_supports_low_power_mode(const gsr_supported_video_c
         case VideoCodec::VP9:         return supported_video_codecs.vp9.low_power;
         case VideoCodec::H264_VULKAN: return supported_video_codecs.h264.low_power;
         case VideoCodec::HEVC_VULKAN: return supported_video_codecs.hevc.low_power; // TODO: hdr, 10 bit
-        case VideoCodec::JPEG:        return false;
-        case VideoCodec::PNG:         return false;
     }
     return false;
 }
@@ -2840,11 +2884,6 @@ static const AVCodec* pick_video_codec(VideoCodec *video_codec, gsr_egl *egl, bo
                 video_codec_f = get_ffmpeg_video_codec(*video_codec, egl->gpu_info.vendor);
             break;
         }
-        case VideoCodec::JPEG:
-        case VideoCodec::PNG: {
-            video_codec_f = get_ffmpeg_video_codec(*video_codec, egl->gpu_info.vendor);
-            break;
-        }
     }
 
     if(!video_codec_auto && !video_codec_f && !is_flv) {
@@ -2906,12 +2945,6 @@ static const AVCodec* pick_video_codec(VideoCodec *video_codec, gsr_egl *egl, bo
                     video_codec_f = get_ffmpeg_video_codec(*video_codec, egl->gpu_info.vendor);
                 break;
             }
-            case VideoCodec::JPEG:
-            case VideoCodec::PNG: {
-                // TODO:
-                //assert(false);
-                break;
-            }
         }
     }
 
@@ -3085,18 +3118,15 @@ static AudioDeviceData create_application_audio_audio_input(const MergedAudioInp
 }
 #endif
 
-static void set_video_codec_for_image_output(const char *filename, VideoCodec *video_codec, const char **video_codec_to_use) {
-    const bool video_codec_auto = strcmp(*video_codec_to_use, "auto") == 0;
+static bool get_image_format_from_filename(const char *filename, gsr_image_format *image_format) {
     if(string_ends_with(filename, ".jpg") || string_ends_with(filename, ".jpeg")) {
-        if(!video_codec_auto)
-            fprintf(stderr, "Warning: expected -k option to be set to 'auto' (or not specified) for jpeg output\n");
-        *video_codec = VideoCodec::JPEG;
-        *video_codec_to_use = "jpeg";
+        *image_format = GSR_IMAGE_FORMAT_JPEG;
+        return true;
     } else if(string_ends_with(filename, ".png")) {
-        if(!video_codec_auto)
-            fprintf(stderr, "Warning: expected -k option to be set to 'auto' (or not specified) for png output\n");
-        *video_codec = VideoCodec::PNG;
-        *video_codec_to_use = "png";
+        *image_format = GSR_IMAGE_FORMAT_PNG;
+        return true;
+    } else {
+        return false;
     }
 }
 
@@ -3739,10 +3769,15 @@ int main(int argc, char **argv) {
 
     const bool is_output_piped = strcmp(filename, "/dev/stdout") == 0;
 
-    set_video_codec_for_image_output(filename, &video_codec, &video_codec_to_use);
-    if(video_codec_is_image(video_codec) && !audio_input_arg.values.empty()) {
-        fprintf(stderr, "Error: can't record audio (-a) when taking a screenshot\n");
-        _exit(1);
+    gsr_image_format image_format;
+    if(get_image_format_from_filename(filename, &image_format)) {
+        if(!audio_input_arg.values.empty()) {
+            fprintf(stderr, "Error: can't record audio (-a) when taking a screenshot\n");
+            _exit(1);
+        }
+
+        capture_image_to_file(filename, window_str, output_resolution, wayland, &egl, image_format, record_cursor, restore_portal_session, portal_session_token_filepath, quality);
+        _exit(0);
     }
 
     AVFormatContext *av_format_context;
@@ -3771,13 +3806,12 @@ int main(int argc, char **argv) {
     const double target_fps = 1.0 / (double)fps;
 
     const bool uses_amix = merged_audio_inputs_should_use_amix(requested_audio_inputs);
-    if(!video_codec_is_image(video_codec))
-        audio_codec = select_audio_codec_with_fallback(audio_codec, file_extension, uses_amix);
+    audio_codec = select_audio_codec_with_fallback(audio_codec, file_extension, uses_amix);
     bool low_power = false;
     const AVCodec *video_codec_f = select_video_codec_with_fallback(&video_codec, video_codec_to_use, file_extension.c_str(), use_software_video_encoder, &egl, &low_power);
 
     const gsr_color_depth color_depth = video_codec_to_bit_depth(video_codec);
-    gsr_capture *capture = create_capture_impl(window_str, output_resolution, wayland, &egl, fps, video_codec, color_range, record_cursor, restore_portal_session, portal_session_token_filepath, color_depth);
+    gsr_capture *capture = create_capture_impl(window_str, output_resolution, wayland, &egl, fps, video_codec_is_hdr(video_codec), color_range, record_cursor, restore_portal_session, portal_session_token_filepath, color_depth);
 
     // (Some?) livestreaming services require at least one audio track to work.
     // If not audio is provided then create one silent audio track.
@@ -3809,20 +3843,32 @@ int main(int argc, char **argv) {
         _exit(1);
     }
     video_frame->format = video_codec_context->pix_fmt;
-    video_frame->width = video_codec_context->width;
-    video_frame->height = video_codec_context->height;
+    video_frame->width = 0;
+    video_frame->height = 0;
     video_frame->color_range = video_codec_context->color_range;
     video_frame->color_primaries = video_codec_context->color_primaries;
     video_frame->color_trc = video_codec_context->color_trc;
     video_frame->colorspace = video_codec_context->colorspace;
     video_frame->chroma_location = video_codec_context->chroma_sample_location;
 
-    int capture_result = gsr_capture_start(capture, video_codec_context, video_frame);
+    gsr_capture_metadata capture_metadata;
+    capture_metadata.width = 0;
+    capture_metadata.height = 0;
+    capture_metadata.fps = fps;
+    capture_metadata.video_codec_context = video_codec_context;
+    capture_metadata.frame = video_frame;
+
+    int capture_result = gsr_capture_start(capture, &capture_metadata);
     if(capture_result != 0) {
         fprintf(stderr, "gsr error: gsr_capture_start failed\n");
         _exit(capture_result);
     }
 
+    video_codec_context->width = capture_metadata.width;
+    video_codec_context->height = capture_metadata.height;
+    video_frame->width = capture_metadata.width;
+    video_frame->height = capture_metadata.height;
+
     gsr_video_encoder *video_encoder = create_video_encoder(&egl, overclock, color_depth, use_software_video_encoder, video_codec);
     if(!video_encoder) {
         fprintf(stderr, "Error: failed to create video encoder\n");
@@ -3849,9 +3895,7 @@ int main(int argc, char **argv) {
 
     gsr_color_conversion_clear(&color_conversion);
 
-    if(video_codec_is_image(video_codec)) {
-        open_video_image(video_codec_context);
-    } else if(use_software_video_encoder) {
+    if(use_software_video_encoder) {
         open_video_software(video_codec_context, quality, pixel_format, hdr, color_depth, bitrate_mode);
     } else {
         open_video_hardware(video_codec_context, quality, very_old_gpu, egl.gpu_info.vendor, pixel_format, hdr, color_depth, bitrate_mode, video_codec, low_power);
@@ -3941,8 +3985,6 @@ int main(int argc, char **argv) {
     if(replay_buffer_size_secs == -1) {
         AVDictionary *options = nullptr;
         av_dict_set(&options, "strict", "experimental", 0);
-        if(video_codec_is_image(video_codec))
-            av_dict_set(&options, "update", "true", 0);
         //av_dict_set_int(&av_format_context->metadata, "video_full_range_flag", 1, 0);
 
         int ret = avformat_write_header(av_format_context, &options);
@@ -4192,7 +4234,6 @@ int main(int argc, char **argv) {
 
     double last_capture_seconds = record_start_time;
     bool wait_until_frame_time_elapsed = false;
-    const bool is_image_output = video_codec_is_image(video_codec);
 
     while(running) {
         const double frame_start = clock_get_monotonic_seconds();
@@ -4272,7 +4313,7 @@ int main(int argc, char **argv) {
 
             // TODO: Dont do this if no damage?
             egl.glClear(0);
-            gsr_capture_capture(capture, video_frame, &color_conversion);
+            gsr_capture_capture(capture, &capture_metadata, &color_conversion);
             gsr_egl_swap_buffers(&egl);
             gsr_video_encoder_copy_textures_to_frame(video_encoder, video_frame, &color_conversion);
 
@@ -4300,10 +4341,6 @@ int main(int argc, char **argv) {
                     // TODO: Move to separate thread because this could write to network (for example when livestreaming)
                     receive_frames(video_codec_context, VIDEO_STREAM_INDEX, video_stream, video_frame->pts, av_format_context,
                         replay_start_time, frame_data_queue, replay_buffer_size_secs, frames_erased, write_output_mutex, paused_time_offset);
-                    if(is_image_output) {
-                        running = 0;
-                        break;
-                    }
                 } else {
                     fprintf(stderr, "Error: avcodec_send_frame failed, error: %s\n", av_error_to_string(ret));
                 }
@@ -4409,7 +4446,7 @@ int main(int argc, char **argv) {
     gsr_damage_deinit(&damage);
     gsr_color_conversion_deinit(&color_conversion);
     gsr_video_encoder_destroy(video_encoder, video_codec_context);
-    gsr_capture_destroy(capture, video_codec_context);
+    gsr_capture_destroy(capture);
 #ifdef GSR_APP_AUDIO
     gsr_pipewire_audio_deinit(&pipewire_audio);
 #endif
diff --git a/src/utils.c b/src/utils.c
index 61ca856..82c51b5 100644
--- a/src/utils.c
+++ b/src/utils.c
@@ -635,9 +635,12 @@ static VADisplay video_codec_context_get_vaapi_display(AVCodecContext *video_cod
 }
 
 bool video_codec_context_is_vaapi(AVCodecContext *video_codec_context) {
+    if(!video_codec_context)
+        return false;
+
     AVBufferRef *hw_frames_ctx = video_codec_context->hw_frames_ctx;
     if(!hw_frames_ctx)
-        return NULL;
+        return false;
 
     AVHWFramesContext *hw_frame_context = (AVHWFramesContext*)hw_frames_ctx->data;
     AVHWDeviceContext *device_context = (AVHWDeviceContext*)hw_frame_context->device_ctx;