Sponsor: Thermaltake SWAFAN EX 120 3-Pack on Amazon https://geni.us/ardZR
This benchmark tests Counter-Strike 2, the successor to CSGO (Global Offensive), for GPU performance across a few maps and a lot of video cards. One of our primary areas of interest for this game has been frametime pacing. Although not too bad overall on our systems, the occasional hitches we did encounter were often large or noticeable even as one-offs. We spend a bit of time exploring those, but also talk about the difference of test environments with smoke grenades, incendiary grenades, or just firefights. We ultimately settled on a benchmark scene that involved smoke grenades for about half the benchmark, with a mix of traversal and firefights for the other half. This represents the average player's round pretty well. Testing included cards like the NVIDIA RTX 4090, RTX 4080, AMD RX 7900 XTX, 7800 XT, and plenty of prior generation cards, like the GTX 1650, GTX 1060, and so on. Intel Arc was also tested for the A750 and A770.
PATCH UPDATE & QUICK TEST: Valve has posted a lot of patches since this video. A couple days after we published, they posted a patch that's meant to address some of the shader compile or frametime spiking ("hitching") behavior on NVIDIA cards specifically. We ran some new numbers against the update and there's improvement in a few places; notably, in instances where we experienced massive stutters on NVIDIA due to (what we suspect) are shader compiling issues, it looks like they're improving. Our RTX 4070 was affected by this in our 1080p testing. AMD was also affected generally. We just ran the 4070 and 4090 through with the new patch and did not encounter the same stutters that hurt the average. The 4070 is now about 8% ahead of the 3070 (as this was effectively an RNG of if the game ran well, it was occasionally ahead and sometimes massively behind last data set, but now is consistently slightly ahead). That was either an error or a bug in the game, our data suggests a bug but we unfortunately don't seem to have a way to run the old version (that we know of - let us know if there is one) to validate against, so we just have to run new numbers and then make educated estimates against prior versions. Valve noted that the frametime pacing behavior is still present on AMD to at least some extent. They claim they're working with AMD, so we assume that means a potential driver update. We'll run everything fresh once Valve has some time to slow the daily updates - they are pushing them out so fast that we can't actually complete a round of tests before the next one, and each update could make the data incomparable, so we'll wait til they go to a slower cadence. Fast updates are good - they're fixing stuff. They're just not good for testing a lot of things.
Starting today and for a few weeks, we are giving away 10% of our store revenue to help our local cat shelter Cat Angels to fund their move to a new adoption center! Check our store here: https://store.gamersnexus.net/
Like our content? Please consider becoming our Patron to support us: http://www.patreon.com/gamersnexus
TIMESTAMPS
00:00 - Counter-Strike 2 Benchmarks
01:47 - CS2 vs. CSGO Improvements
03:39 - Research: Maps & Smoke Grenades
05:56 - Frame Pacing Issues
07:08 - Frametimes (7900 XTX)
08:06 - Frametimes (RTX 4060)
08:43 - Frametimes (RTX 4070)
10:52 - 4K GPU Benchmarks in Counter-Strike 2
12:34 - 1440p GPU Benchmarks
13:35 - 1080p CS2 GPU Benchmarks
15:04 - Check Pinned Comment & Conclusion
** Please like, comment, and subscribe for more! **
Links to Amazon and Newegg are typically monetized on our channel (affiliate links) and may return a commission of sales to us from the retailer. This is unrelated to the product manufacturer. Any advertisements or sponsorships are disclosed within the video ("this video is brought to you by") and above the fold in the description. We do not ever produce paid content or "sponsored content" (meaning that the content is our idea and is not funded externally aside from whatever ad placement is in the beginning) and we do not ever charge manufacturers for coverage.
Follow us in these locations for more gaming and hardware updates:
t: http://www.twitter.com/gamersnexus
f: http://www.facebook.com/gamersnexus
w: http://www.gamersnexus.net/
Host, Writing, Testing, Editing: Steve Burke
Testing, Bench Course Match: Mike Gaglione
Testing: Jeremy Clayton
This benchmark tests Counter-Strike 2, the successor to CSGO (Global Offensive), for GPU performance across a few maps and a lot of video cards. One of our primary areas of interest for this game has been frametime pacing. Although not too bad overall on our systems, the occasional hitches we did encounter were often large or noticeable even as one-offs. We spend a bit of time exploring those, but also talk about the difference of test environments with smoke grenades, incendiary grenades, or just firefights. We ultimately settled on a benchmark scene that involved smoke grenades for about half the benchmark, with a mix of traversal and firefights for the other half. This represents the average player's round pretty well. Testing included cards like the NVIDIA RTX 4090, RTX 4080, AMD RX 7900 XTX, 7800 XT, and plenty of prior generation cards, like the GTX 1650, GTX 1060, and so on. Intel Arc was also tested for the A750 and A770.
PATCH UPDATE & QUICK TEST: Valve has posted a lot of patches since this video. A couple days after we published, they posted a patch that's meant to address some of the shader compile or frametime spiking ("hitching") behavior on NVIDIA cards specifically. We ran some new numbers against the update and there's improvement in a few places; notably, in instances where we experienced massive stutters on NVIDIA due to (what we suspect) are shader compiling issues, it looks like they're improving. Our RTX 4070 was affected by this in our 1080p testing. AMD was also affected generally. We just ran the 4070 and 4090 through with the new patch and did not encounter the same stutters that hurt the average. The 4070 is now about 8% ahead of the 3070 (as this was effectively an RNG of if the game ran well, it was occasionally ahead and sometimes massively behind last data set, but now is consistently slightly ahead). That was either an error or a bug in the game, our data suggests a bug but we unfortunately don't seem to have a way to run the old version (that we know of - let us know if there is one) to validate against, so we just have to run new numbers and then make educated estimates against prior versions. Valve noted that the frametime pacing behavior is still present on AMD to at least some extent. They claim they're working with AMD, so we assume that means a potential driver update. We'll run everything fresh once Valve has some time to slow the daily updates - they are pushing them out so fast that we can't actually complete a round of tests before the next one, and each update could make the data incomparable, so we'll wait til they go to a slower cadence. Fast updates are good - they're fixing stuff. They're just not good for testing a lot of things.
Starting today and for a few weeks, we are giving away 10% of our store revenue to help our local cat shelter Cat Angels to fund their move to a new adoption center! Check our store here: https://store.gamersnexus.net/
Like our content? Please consider becoming our Patron to support us: http://www.patreon.com/gamersnexus
TIMESTAMPS
00:00 - Counter-Strike 2 Benchmarks
01:47 - CS2 vs. CSGO Improvements
03:39 - Research: Maps & Smoke Grenades
05:56 - Frame Pacing Issues
07:08 - Frametimes (7900 XTX)
08:06 - Frametimes (RTX 4060)
08:43 - Frametimes (RTX 4070)
10:52 - 4K GPU Benchmarks in Counter-Strike 2
12:34 - 1440p GPU Benchmarks
13:35 - 1080p CS2 GPU Benchmarks
15:04 - Check Pinned Comment & Conclusion
** Please like, comment, and subscribe for more! **
Links to Amazon and Newegg are typically monetized on our channel (affiliate links) and may return a commission of sales to us from the retailer. This is unrelated to the product manufacturer. Any advertisements or sponsorships are disclosed within the video ("this video is brought to you by") and above the fold in the description. We do not ever produce paid content or "sponsored content" (meaning that the content is our idea and is not funded externally aside from whatever ad placement is in the beginning) and we do not ever charge manufacturers for coverage.
Follow us in these locations for more gaming and hardware updates:
t: http://www.twitter.com/gamersnexus
f: http://www.facebook.com/gamersnexus
w: http://www.gamersnexus.net/
Host, Writing, Testing, Editing: Steve Burke
Testing, Bench Course Match: Mike Gaglione
Testing: Jeremy Clayton
- Категория
- Экшен
Комментариев нет.