I might just use the latter since youtube will compress it even more than I will.RyviusARC

Yeah, probably should. When converting to something like Xvid or H264, a YV12 and YUY2 source should look pretty much the same when re-encoded. YUY2 just gets converted to YV12 anyway.

The main reason you'd want something more than YV12 is if you were going to be doing any fancy post-processing effects in a video editor. You'd want RGB for the best effect rendering quality. I don't think YV12 vs RGB matters for direct screen footage converted to something like H264, though.. it should end up the same.

edit: by the way, Lagarith YV12 looks slightly better (sharper) than DxtoryVC "Medium" in my comparison. Almost the same, so use whichever runs better for you.

Well I tried FRAPS on non RGB and Dxtory on Medium and FRAPS is crisper around the edges. Edges were slightly blurrier on Dxtory. Only reason I noticed this was that it hides part of the aliasing.RyviusARC

Yup, you're right. See my post from a couple back.

Also, Dxtory YUV24 looks better than non-RGB FRAPS, but it requires insane write speed. Lagarith YUY2 is similarly good quality (better than FRAPS), and it requires lower bitrate than FRAPS non-RGB.. so I would check how that runs for you.

everyone who cannot afford a capture card and uses software for recording..is using FRAPS...why?rhazzy

Because it's been around far longer and is better-known, obviously.

Capture cards suck compared to direct screen caps, by the way.

FRAPS is great, but it's starting to show its age. Dxtory is relatively young, and yet it still manages to be better overall with a proper setup. It's likely that it's only going to get better and better as time goes on, and FRAPS will be left in the dust unless they do a significant revamp of that outdated codebase and add in stuff like multithreading, custom resolutions, and a proper output component that isn't limited to 4GB chunks.

Hmm.. I need to try PlayClaw.

Not sure if anyone cares, but I did some self-observed quality comparisons between FRAPS and Dxtory with Lagarith/DxtoryVC.. rendered in VirtualDub and exported to PNG. Some unexpected results:

FRAPS non-RGB actually seems to be using something like YUV24 (*not* YV12 as I initially thought).. but the colors are not accurate. I don't get it. Even the FRAPS RGB colors (supposedly lossless) are not accurate to the original. They're way brighter/more saturated than the original (comparing to an in-game screencap) destroying a lot of the original color detail. I'm guessing that this actually has something to do with the FRAPS codec not reporting the colorspace properly to the renderer, rather than FRAPS not storing color information properly.. but that's really weird.

Dxtory RGB and Lagarith RGB look EXACTLY like the original game screencap, as expected. FRAPS RGB, again, has inaccurate colors for reasons unclear.

Detail-wise (ignoring color inaccuracy), FRAPS non-RGB looks sharper than Dxtory YUV420 and Lagarith YV12. I expected them to be the same, but FRAPS wins out here.

However, Dxtory YUV24 and Lagarith YUY2 look better than non-RGB FRAPS. Lagarith achieves this better quality at a lower bitrate. Dxtory.. not so much (a lot higher bitrate that FRAPS.)

Would post the screenshots but they're each 2.9MB to 3MB PNG files.

Again, everything gets converted to YV12 when you convert to a lossy format like H264, so I'm really not sure that it matters. I'll do some x264 test encodes later if I get bored.

Boy oh boy...ur the one who dosent get it...WithLagarith @YV12 and DxtoryVC @"Medium" you will get the half of the fps u get by using Fraps "non-RBG"...Do you get it now?Ofc is convenient if ur recording CS or Starcraft...but if ur recording some FPS like BC 2 of BF 3 or any other graphically intensive game u will have a problem...

For example in Rage...With Dxtory(Lagarith @YV12) 1920x1080 i get 21 fps(on the game and also on the file) with Fraps i get 36...Do you get it now?Lets even assume that Dxtory and Fraps have same picture quality...how the hell are u suppose to play a game at 20-21 fps?How are u suppose to make a fluid video at that framerate?

rhazzy

The more likely reason for your issues is that HDD setup in your sig.

Never record to an external drive.

Never record to the OS drive.

If FRAPS works better for you with that setup, great. I'm telling you that this is not the case for my setup.

Wow.. you have some compelling arguments there, rhazzy. :roll:

Quit trying to justify your license purchase to me already, kid. I really don't care which you prefer; I'm just trying to help people understand the pros/cons of each, not engage in the blind/ignorant fanboyism you seem to enjoy reveling in.

I'm just going by what I see on youtube... FRAPS is crisper than Dxtory.realguitarhero5

Ah. Well, YouTube videos are re-encoded by the uploader, uploaded, and then re-encoded again by YouTube. Not something by which you can consistently judge recording quality.. especially when the settings/encoders uploaders use is hugely variable.

I'll have to test this later and upload it to ScreenshotComparison.

FRAPS (non-RGB) vs. Dxtory Medium vs. Lagarith YV12. They should look the same, but I've never actually tested it. Kinda interested now.

Dxtory. Very little framerate loss, but it looks slightly less crisp than FRAPS. However I would rather have my frames than a *slightly* worse looking recording.realguitarhero5

Hmm.. you sure it's not placebo? What settings?

Try some screenshots at different quality settings, saved as BMP or PNG. Dxtory's "Medium" should be pretty much the same as FRAPS non-RGB.

These YV12 formats do not look exactly like the original, because the color storage is lossy. They convert to YV12, which is the same colorspace that any good modern encoder uses, so it doesn't really matter if you use RGB or YV12.. the end-result will look the same once it's in something like H264 (ONLY supports YV12.) It's only the initial recording that will appear slightly different.

I'll have to do some comparisons later.

As for FRAPS RGB and Dxtory RGB, they should be 1:1 storage of the original game footage, and therefore identical.

Seems like Dxtory doesn't work with some games. I tried The Last Remnant and it wouldn't record. So FRAPS is still a must have for some games because Dxtory doesn't work with all games.RyviusARC

Interesting. I've heard that Dxtory sometimes doesn't work with really old DirectX/OGL games; never heard of it not working with something that recent though.

Haven't run across any incompatibilities yet, but I'll keep FRAPS handy.

FRAPS is the better choice bacause of the higher frame rates that u get ingame when recording...

Unless u want to have 1:1 quality useing a "lossless codec" and record like its a slide show...rhazzy

FRAPS /w RGB checked *is* "lossless."

FRAPS non-RGB is not lossless for purposes of recording game footage. Lagarith @YV12 and DxtoryVC @"Medium" are technically the same thing as FRAPS non-RGB.. they are only "lossless" post-colorspace-conversion (i.e. NOT lossless for purposes of recording game footage.) The difference between them is that Lagarith is more efficient than the others. Get it..?

Lagarith also seems to play more nicely with video editing programs, which is a nice bonus.

"if you can show me a video of a game that is recorded with dxtory which is better in quality and performance than one recorded with fraps"rhazzy

I didn't answer because the question is dumb and displays a basic lack of understanding of digital video encoding. To elaborate:

If they're recorded with the same colorspace with a lossless encoder, transcoded with x264 with the same settings on the same system, uploaded to YouTube under the same conditions, then there will be no perceivable quality differences. FRAPS can't record higher quality than Dxtory, and vice versa. They are the same quality-wise, if you have everything configured properly.

Performance-wise, Dxtory should perform better than FRAPS with a similar-quality configuration. In-game framerate will not be locked to the rate of recording, and Dxtory supports multi-threading and file outputs greater than 4GB. You can't show stuff like a higher playing framerate in a video because the video is recording slower than the game is drawing frames.