I am fairly certain it is the XTX variant, but you might want to contact Sapphire directly to ask.Some amazing prices there for AIB 6900XT's. Is the Toxic the XTXH version?
Had I not purchased a 3080Ti last month, I would have bought that.
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
I am fairly certain it is the XTX variant, but you might want to contact Sapphire directly to ask.Some amazing prices there for AIB 6900XT's. Is the Toxic the XTXH version?
Had I not purchased a 3080Ti last month, I would have bought that.
Some amazing prices there for AIB 6900XT's. Is the Toxic the XTXH version?
Yep to each their own and all that.Not a fan of that gold with the silver. The blue on the front of the box with the gold would have been way better.
Still, if the 7x00 cards maintain an AIO I would love to get one next gen. No hurry this time though!
This one does look cool if your into themed rigs you can show off.
If only FM could offload some of that parallel compute to the GPU eh Matt?![]()
Was playing around on Doom Eternal earlier, captured some footage with RT on and off. I disable RT at around 8 minutes 10 seconds in. Will convert to 4K quality at some point.
I thought @Nexus18 and and @ChocAndGray might find this interesting as they were discussing occasional spikes in frame time latency in Doom Eternal.
Might get a 3080 if a good deal comes up tbh mate.HU mentioned it was one of the games they seen struggle on the 30 series cards when out of vram, not played it myself, shame you offloaded the 3090 matt!
Uploading some from my end of same area and same settings, all in all, my fps looks higher majority of the time, at the start frame latency is a bit hitchy but smooths out nicely but yours looks to have zero spikes. What's your MSI AB polling rate set to?
Somehow, your setup always seem to get better results than any reviewers benches too
![]()
FPS on my 3080 seems about the the same as theirs (mine is .825MHz @ 1815MHz though)
Any ideas why the 6800xt is showing some bad spikes in other peoples footage e.g.
![]()
I watched that video and honestly it looks like a possible local system issue.
This is the problem with taking random videos on YouTube IMO, we have no idea if their system is even stable or what state their OS is in.
Might get a 3080 if a good deal comes up tbh mate.
My rate is 500ms.
Yep I would expect your FPS to be higher with RT on, I'm sure that will change if you toggle it off like I did though half way through.
My GPU is obviously overclocked and 6900 XTs will overclock very well if cooled appropriately. Adding another 15-20%+ over a stock 6900 XT is easily possible due to the low power draw so plenty of headroom there.
I watched that video and honestly it looks like a possible local system issue.
What I noticed is that Jansen user looks to have manually overclocked the CPU to 4.65Ghz. However, in parts of the benchmark all but two of his CPU cores are operating at the base frequency, which is 3.8Ghz. See specs here.
![]()
By running an all core overclock he is reducing system performance as if he had left the CPU at stock, or even just used Precision Boost Overdrive, many of his CPU cores could be operating up to or near 5Ghz instead of 3.8Ghz-4.65Ghz.
That's a problem and should not be happening. Is that definitely the cause? Don't know the game is not that heavy on the CPU so perhaps not but it's a red flag to me that something is not right with his config.
If it is not that, then my next best guess would be it's an old driver/bios/game version issue. That video was uploaded in Jan 2021 so might be that.
This is the problem with taking random videos on YouTube IMO, we have no idea if their system is even stable or what state their OS is in.
Come on I'm not sure how you can say that with a straight face given your stance on FC6.Local system issue...... Random videos on youtube..... Funny how that logic doesn't always apply in other cases eh....![]()
Yeah it's hard to find newer footage, even bangperbucks 6900xt shows some spikes from 9 months ago:
And yes no doubt without RT, 6900xt would be higher although if going by computerbase.de results, not a whole lot surprisingly:
![]()
Any advice/tips for my 5600x, I'm currently using PBO with a -30 on the curve across all cores, which so far has proved rather good especially on the temp front but as mentioned I do notice these kind of spikes every now and then in various games regardless of resolution or vram usage etc.
Hopefully my video will be done soon so you can see cpu core speed etc.
Come on I'm not sure how you can say that with a straight face given your stance on FC6.
I've no problem with Computerbases results tbh, I know they are good people with their testing just like PcGamesHardware.
This is why I always put videos up, even with the FPS cost from recording. Nothing says it better like a video with an overlay to back up my claim.
Re Bangperbuck, maybe it was a driver/game issue that was resolved, not sure.
If you can achieve -30 on all cores then you must have a very good sample so not sure you can improve on it much. I can only run a few cores at -30. My best core will only do -2 or so.
I don't think your spikes are related to the CPU, but I'll have a look at your video when it's ready.
If I had to guess why you both get frame time graph spikes I'd suggest it is probably a driver issue. That said, I know 10GB is borderline for Doom Eternal, so it could be related to that also.![]()
I have Game Bar off, though as I'm using Win 11 I don't think you can truly disable it like you can on Windows 10.In the sense, when I say it's a possible system issue then get laughed at yet here we are...![]()
Yeah it's very weird and always hard to tell as it isn't every game, obviously most unreal engine 4 games have it due to the way the level/map is loaded or shader cache or whatever it is that DF always mentions but I can notice it in non UE games too i.e. BF 5 (where I have everything on low/off at 1440p). It's not really a big issue as it doesn't happen regularly and as mentioned, not in "every" game either, just when you mentioned the guys cpu, I was wondering if that could be reason on my end possibly as I know these ryzen cpus can be a bit weird in their behaviour when it comes to OC/UV and RAM timings + speed. My RAM is just crucial 3200MHz (dual rank with xmp enabled)
What about hardware acceleration, game bar etc. Are they all off?
BTW, iirc, choc has a 3090 so defo won't be vram on his end.
I have Game Bar off, though as I'm using Win 11 I don't think you can truly disable it like you can on Windows 10.
I do have hardware acceleration enabled in my browsers, but I don't have those running when I record videos. Just the game and my monitoring apps/drivers.
I thought that last question was a bit odd.Ah sorry keep forgetting microsoft changed the name, used to be called hardware acceleration, now it is called game mode:
![]()
I keep game bar and game mode on but have read it can cause some issues but then others say it helps them....
1080P version done, 4k still processing though:
If you can point me towards some (reflections) near the start of the game, I'll take some screenshots and or do another video.Nice one. Certainly a good comparison. Yeah it's a perfect size my OSD for 3440x1440 but not so much 4k
No scenes with some reflections?![]()
As was noted in another thread, it's quite interesting seeing the power consumption to performance with RT titles too, in the sense, RDNA 2 is definitely better on the whole but when it comes to RT, this is were ampere is quite a bit better.
I'm surprised FSR hasn't been added to this yet, can't recall but does FSR work with vulkan?
BTW, how does my CPU behaviour look? Normal enough?