Software for recording games?

most people set it up wrong or dont have a spare drive for the files that can be big. the thing is i use all progs. the main i use is fraps because sorry its better than shadow play quality wise.

thing is people say about the size of raw files. well you shouldnt be recording on same drive as you play for a start and how much is a normal mech drive 30-40 quid for a terrabyte.

so fraps is fine.just set up right and have different drives to record on.you should do that regardless of software you use.

the best are dxtory/fraps .

at end most of it comes down to user interface and what you feel happy with. then quality and performance.

+1

The quality of fraps does seem to be among the better examples out there.


I also still think Afterburner with VFW compression + x264 / lagarith lossless codec is solid as well.
 
Last edited:
I'll be doing a comparison of various recording methods for teh lulz, so could you guys that have solid 1080p/60 settings for Afterburner w/VFR/Lagarith, OBS, Dxtory and fraps please post screenshots of how you have them set up please?

Right now, I'm still getting the best results from Mirillis, so I'd like to know if I'm doing something wrong with the others.
I'll be posting links to zipped capture footage, rather than uploading to youtube for the purpose of direct comparison without YT compression shenanigans.

Cheers.
 
Let's hope so :)

I own that too, problem I find is that it drops FPS quite a bit in comparison to recording at the same quality on Shadowplay. That's when measured through Steam or other third party FPS tools.

Also had numerous problems with MA over the past couple of years, audio tracks not importing properly, it recording at half speed so the clip is played back faster etc. Support frankly is very poor.

It can drop fps a little more than others, yes.
Right now though, I at least seem to have no problem capturing at 1080p/60 fwiw, though I do recall a couple of earlier builds being a bit ropey here and there.

Quality is noticeably better than Shadowplay using NVENC H.264 IMO.
 
If you want quick and convenient recording (and have an nVidia card) then ShadowPlay isn't a bad choice, especially as recording performance is very quick.

However, if you want the best possible quality then you really need to be looking at a software-based recorder (e.g. Dxtory) and a fast, lossless codec (lagarith being one of the top choices).

Transcode your files using x264 for long-term storage or uploading to YouTube.

I compared H.264 encodings using QuickSync and nVidia's solutions a while back. While they might be quick, the overall quality of x.264 output will generally be much higher and the file sizes even smaller (depending on the encoder settings).
 
So anyone got any 1080p/60 settings that they would recommend along with their program of choice?

I only ask so as to avoid the 'you're doing it wrong' comments that will inevitably happen if I post results that don't live up to what people are achieving or expecting.
 
Well, I'm relatively new to all this, but for MSI Afterburner this is what I'm using at the mo (sometimes use lagarith too) :

2lnibf5.png

No idea why the video capture key disappeared. Need to reset it. :)
 
Last edited:
Ok, so here we are :)

I've recorded a lap from a replay of a recent league race in Assetto Corsa.
Steam and the game are running from an SSD and recordings are made to a separate SSD.

I haven't included Lagarith recordings from either Afterburner or Dxtory because they resulted in file sizes around the 8GB mark for the just under 2 minute vids.
I didn't feel that quality was notably better than the best of the others anyway tbh - certainly not enough to justify the file size.

I also didn't include Fraps, as you can only record 29secs unless you buy it. That 29sec vid was still a 2.4GB file and as above, not worth it IMO.
There aren't many settings in there, so if setting it to 1080p @60fps isn't doing it right, oh well.

OBS isn't here because I couldn't be bothered to trawl around looking for good settings/codecs etc, or go through a lengthy trial and error session.

Afterburner NVENC default settings LINK 90MB
Given the very small filesize, it's not surprising that this is the worst of the lot.
It might be good for other types of games though. Perhaps something that remains fairly static vs lots of panning shots.

DXtory VFW x264 LINK 602MB
fnf62kZ.png
Mirillis NVENC LINK 429MB
nVgOuz1.png
^^These two had the biggest performance hit, or at least the lowest minimums.
In the case of Mirillis, I'm not sure if that had anything to do with locking fps to whatever recording rate you've set it to.

Afterburner NVENC MAX settings LINK 718MB
ok0Y55Y.png
Afterburner VFW x264 AVI LINK 554MB
FI7cu7d.png
Afterburner VFW x264 MKV LINK 637MB
Same settings as above and performance hit is the same

Shadowplay MAX settings LINK 697MB

The above are all much of a muchness, though you may have better luck finding any obvious differences between them.
I take back what I said about Shadowplay though. It's on par.

I haven't covered every base here. There are alternate recording codecs in Mirillis (avi/mp4) and many options to play with in most of the other software.

What would I choose? Probably Afterburner VFW/x264.

ED: Don't go by the web player. Download the clips to see the actual quality.
 
Last edited:
Good work, Petey.

As I understand it, MKV is superior in quality to AVI. I also would agree that out of those, Afterburner is probably the best overall, though there's not that much in it.

Out of the many I've looked at, D3DGear was another which seemed to be among the better ones.
 
Last edited:
Yeah, it's all pretty close. I'll have a look at D3DGear too. Cheers.
Another thing to note is ease of use/setup. You can't beat Shadowplay/Mirillis for that.

Afterburner/Dxtory require a bit of research, downloading extra stuff and tweaking, but it does show that you can get results that are just as good or better with free/vendor agnostic capture software vs paid apps.

Btw, if anyone's clicking those links and making comparisons via the web player, you'll find that actually downloading the file will give the true result as the web player isn't doing 60fps playback and is probably compressing them.
 
Last edited:
Back
Top Bottom