• Competitor rules

    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.

Nvidia vs AMD vram usage?

Soldato
Joined
25 Nov 2011
Posts
20,679
Location
The KOP
Just got a 480 vs 1060 pop up in my YouTube feed and something I noticed again is just how much difference they is between the two running same setting Amd is around 2GB and nvidia 4GB
Anyone have any idea what is going on here?

Here link sorry can't add to post correct on phone so only mobile links.
https://youtu.be/C9QhOOGm_2k

Seen this on a Farcry primal video other day I'll find link later, where Vram usage was much lower on the AMD GPU than Nvidia running same settings.
 
Just got a 480 vs 1060 pop up in my YouTube feed and something I noticed again is just how much difference they is between the two running same setting Amd is around 2GB and nvidia 4GB
Anyone have any idea what is going on here?

Here link sorry can't add to post correct on phone so only mobile links.
https://youtu.be/C9QhOOGm_2k

Seen this on a Farcry primal video other day I'll find link later, where Vram usage was much lower on the AMD GPU than Nvidia running same settings.

think it just depends on the game if it has nvidia game works like tress fx etc etc
this video shows all the cards within 300mb of each other
 
Problem is it could be quite complex reasons why - i.e. whether the VRAM is actually used or just lazy garbage collection or caching, etc. nVidia also do some stuff with complied shader caching which can also increase VRAM use but also results in less stutter in some situations compared to normal.

EDIT: Wow 480 gets smoked in that GTA V video.
 
Last edited:
Problem is it could be quite complex reasons why - i.e. whether the VRAM is actually used or just lazy garbage collection or caching, etc. nVidia also do some stuff with complied shader caching which can also increase VRAM use but also results in less stutter in some situations compared to normal.

EDIT: Wow 480 gets smoked in that GTA V video.

Be interested what would happen if they selected shader cache in Radeon settings.

Yeah GTA5 seems to favour Nvidia quite abit, still though 480 is handling that sweet 60fps though.
 
Blimey, the 970 is smashing up the 480 in that video big time :eek:

780ti is giving the 970 a fair kicking at well - I suspect it is due to the amount of primitive geometry onscreen in those kind of games - one of the reasons I mostly stick to the 780 is that a large part of my gaming these days is competitive FPS with "lowpro" settings where Kepler just dishes it out compared to the newer cards that are optimised around shader performance.
 
To be honest, his (DudeRandom84's) testing methodology isn't scrupulous enough.

Tinfoil hat moment:

He could have aggressive pre-allocation set as the memory policy for Nvidia, by some weird force of habit.

He might have enabled "High Quality" texture filtering in NVCP, which can raise VRAM usage quite a lot in some games as it disables all optimizations.

In the OP video, the RX 480 is throttling hard. The core clock is all over the place. In the next video, it's 100% stable at 1266MHz.

It's also not mentioned if it's a 4GB or 8GB 480, as the settings screen is clearly recorded on the 6GB 1060 in the OP video.

Don't get me wrong, I appreciate the work he puts in and he obviously likes buying GPUs but more information would be nice :p
 
780ti is giving the 970 a fair kicking at well - I suspect it is due to the amount of primitive geometry onscreen in those kind of games - one of the reasons I mostly stick to the 780 is that a large part of my gaming these days is competitive FPS with "lowpro" settings where Kepler just dishes it out compared to the newer cards that are optimised around shader performance.

Yer, just watched again and no gimping on the 780Ti there anyways :p
 
To be honest, his (DudeRandom84's) testing methodology isn't scrupulous enough.

Tinfoil hat moment:

He could have aggressive pre-allocation set as the memory policy for Nvidia, by some weird force of habit.

He might have enabled "High Quality" texture filtering in NVCP, which can raise VRAM usage quite a lot in some games as it disables all optimizations.

In the OP video, the RX 480 is throttling hard. The core clock is all over the place. In the next video, it's 100% stable at 1266MHz.

It's also not mentioned if it's a 4GB or 8GB 480, as the settings screen is clearly recorded on the 6GB 1060 in the OP video.

Don't get me wrong, I appreciate the work he puts in and he obviously likes buying GPUs but more information would be nice :p

More information would definitely be nice, it's the only real downside I can give Digital Foundry they don't give much information in there videos either about cards type or clock speeds.
One thing all YouTube channels should definitely aim for if they into this video recording is external capture to completely remove that bottleneck.

OP video am not sure what he uses for recording? But I do know some games are effected much more depending on recording software.
 
Back
Top Bottom