Jump to content

Anyone know when GTX 970/980 with 8GB vram may come out?

ChineseChef
Go to solution Solved by Glenwing,

Scenario question:  If the card can only effectively use 4GB, but they give it 8GB, and the game tries to use 8GB, will the card use it effectively?  Or will it flounder?

 

Also, do you have any links or directions to look for a good explanation about "effective" use of VRAM in gfx cards?  I understand addressable limits and such, but not why a card couldn't use more RAM.  Unless it is simply that the card only has enough processing power to use X amount.

 

It's a matter of memory bandwidth, the GPU buffers frames in the memory continuously and pulls them out to send to the display continuously, so there is a huge amount of information transfer going in and out across the memory bus at all times.  The memory bus has a certain maximum bandwidth (bus width times effective frequency, 256-bit x 7GHz = 1.792Tbits/s = 224GB/s) and so it can only facilitate so much communication.  If the memory bandwidth is being maxed, then the GPU simply won't be able to store or retrieve frames any faster than the memory bus is capable of handling per second, which translates into a framerate limitation.  Any extra frames that the GPU is capable of generating are simply lost, as it has nowhere to buffer the frame since the memory bus cannot transfer any more data to the memory than it already is.  If you are running at very high resolution with very large frames to buffer, and you are limited by memory bandwidth, you'll have a lower framerate than running at lower resolution (again, assuming everything is limited by memory bandwidth).  The memory bus has a maximum data transfer rate, and larger frames contain more data, so not as many of them will fit across the bus per second. At this point adding extra capacity won't matter, because the GPU can't fill any more of the VRAM capacity with frames, data transfer to the memory cannot go any faster.  It's like a congested road, it doesn't matter how big the parking lot at the other end is, if the road only has one lane, there is a maximum limit to how many cars can come and go in an hour (assuming everyone respects the speed limit).  So... no point in expanding that parking lot beyond a certain size, until you add some lanes (wider bus) or crank up the speed limit (higher frequency).

 

This is very application dependent though; if you are using the GPU as a GPU, the memory is used for frame buffering, and the above applies.  Things are a bit different if you are using the GPU as a co-processor for accelerating CPU computations (in a word, GPGPU).  In this type of usage it is not using the VRAM for storing frames and retrieving them constantly and as fast as possible.  It does store a large amount of information that it needs for doing its work, but in this case much more memory can be utilized effectively, because it is not trying to access massive amounts of it at once, the data transfer and communication is comparatively very small compared to frame buffering.  It just stores a lot of information, and accesses a little bit at a time.  Just what it needs at any given point for its computations.  That's why for example you see a Quadro K6000 with 12GB of VRAM on a 384-bit 6GHz bus (288GB/s) and it uses all of it no problem.

 

Of course, you don't even need to go into memory bandwidth for this situation, it's a simple matter of GPU raw computation power.  A GTX 980 is not going to be generating enough frames in gaming to fill up 8GB of VRAM anyway, and the GPGPU capabilities are severely crippled in GeForce cards (artificially) so it's a very bad option for that.  Maybe it makes sense when you think about two-way or three-way SLI (since the memory capacity is not combined) but then again, you run into bandwidth issues in that case...

All of the more recent cards have eventually released a card with double VRAM.  I am under no delusion it will come out soon, but was just curious when everyone thought larger VRAm cards will start being released.  With the current consoles having "8GB" of VRAM, and game devs starting to use more vram and ram in general, it may actually be useful for more than edge cases.

Link to comment
Share on other sites

Link to post
Share on other sites

the 780 w/ 6GB came out quite a while after the 780 release.

I wouldnt be surprised if you had to wait a few months 

Are you new? Please read CoC  Before posting! PLEASE SELECT 'AUTOMATIC' FOR FONT COLOR FOR US DARK THEME USERS (Only for dark shades)If you can read this, it means you need to change to NIGHT THEME (bottom left above LMG logo) Please follow your threads and Quote people when replying to them. Mark your problem as solved if answered. Don't spam, nobody likes a spammer. DO NOT QUOTE IMAGES! BE NICE!

Link to comment
Share on other sites

Link to post
Share on other sites

Graphics cards have limits to the amount of VRAM they can effectively make use of in games.  I have no doubt they will release double-VRAM cards as usual but I also doubt it will even be worth thinking about.  They don't do it because it's useful, they do it because someone somewhere will pay extra money for it.

Link to comment
Share on other sites

Link to post
Share on other sites

I doubt the 970 would really need more than the 4gb it has.

Specs: CPU - Intel i7 8700K @ 5GHz | GPU - Gigabyte GTX 970 G1 Gaming | Motherboard - ASUS Strix Z370-G WIFI AC | RAM - XPG Gammix DDR4-3000MHz 32GB (2x16GB) | Main Drive - Samsung 850 Evo 500GB M.2 | Other Drives - 7TB/3 Drives | CPU Cooler - Corsair H100i Pro | Case - Fractal Design Define C Mini TG | Power Supply - EVGA G3 850W

Link to comment
Share on other sites

Link to post
Share on other sites

Not really on topic, but whats the next generation after maxwell 970's ?

Link to comment
Share on other sites

Link to post
Share on other sites

Graphics cards have limits to the amount of VRAM they can effectively make use of in games.  I have no doubt they will release double-VRAM cards as usual but I also doubt it will even be worth thinking about.  They don't do it because it's useful, they do it because someone somewhere will pay extra money for it.

 

Scenario question:  If the card can only effectively use 4GB, but they give it 8GB, and the game tries to use 8GB, will the card use it effectively?  Or will it flounder?

 

Also, do you have any links or directions to look for a good explanation about "effective" use of VRAM in gfx cards?  I understand addressable limits and such, but not why a card couldn't use more RAM.  Unless it is simply that the card only has enough processing power to use X amount.

Link to comment
Share on other sites

Link to post
Share on other sites

Vram is not really that expensive so the oems usually produce the cards with the optimal capacity vram. Double vram is no longer useful unless you are running some insane dual 4k setup. 

Link to comment
Share on other sites

Link to post
Share on other sites

Scenario question:  If the card can only effectively use 4GB, but they give it 8GB, and the game tries to use 8GB, will the card use it effectively?  Or will it flounder?

 

Also, do you have any links or directions to look for a good explanation about "effective" use of VRAM in gfx cards?  I understand addressable limits and such, but not why a card couldn't use more RAM.  Unless it is simply that the card only has enough processing power to use X amount.

 

It's a matter of memory bandwidth, the GPU buffers frames in the memory continuously and pulls them out to send to the display continuously, so there is a huge amount of information transfer going in and out across the memory bus at all times.  The memory bus has a certain maximum bandwidth (bus width times effective frequency, 256-bit x 7GHz = 1.792Tbits/s = 224GB/s) and so it can only facilitate so much communication.  If the memory bandwidth is being maxed, then the GPU simply won't be able to store or retrieve frames any faster than the memory bus is capable of handling per second, which translates into a framerate limitation.  Any extra frames that the GPU is capable of generating are simply lost, as it has nowhere to buffer the frame since the memory bus cannot transfer any more data to the memory than it already is.  If you are running at very high resolution with very large frames to buffer, and you are limited by memory bandwidth, you'll have a lower framerate than running at lower resolution (again, assuming everything is limited by memory bandwidth).  The memory bus has a maximum data transfer rate, and larger frames contain more data, so not as many of them will fit across the bus per second. At this point adding extra capacity won't matter, because the GPU can't fill any more of the VRAM capacity with frames, data transfer to the memory cannot go any faster.  It's like a congested road, it doesn't matter how big the parking lot at the other end is, if the road only has one lane, there is a maximum limit to how many cars can come and go in an hour (assuming everyone respects the speed limit).  So... no point in expanding that parking lot beyond a certain size, until you add some lanes (wider bus) or crank up the speed limit (higher frequency).

 

This is very application dependent though; if you are using the GPU as a GPU, the memory is used for frame buffering, and the above applies.  Things are a bit different if you are using the GPU as a co-processor for accelerating CPU computations (in a word, GPGPU).  In this type of usage it is not using the VRAM for storing frames and retrieving them constantly and as fast as possible.  It does store a large amount of information that it needs for doing its work, but in this case much more memory can be utilized effectively, because it is not trying to access massive amounts of it at once, the data transfer and communication is comparatively very small compared to frame buffering.  It just stores a lot of information, and accesses a little bit at a time.  Just what it needs at any given point for its computations.  That's why for example you see a Quadro K6000 with 12GB of VRAM on a 384-bit 6GHz bus (288GB/s) and it uses all of it no problem.

 

Of course, you don't even need to go into memory bandwidth for this situation, it's a simple matter of GPU raw computation power.  A GTX 980 is not going to be generating enough frames in gaming to fill up 8GB of VRAM anyway, and the GPGPU capabilities are severely crippled in GeForce cards (artificially) so it's a very bad option for that.  Maybe it makes sense when you think about two-way or three-way SLI (since the memory capacity is not combined) but then again, you run into bandwidth issues in that case...

Link to comment
Share on other sites

Link to post
Share on other sites

snip....Awesome explanation

 

That is probably one of the best simple explanations I have read in a long time. 

 

So with games having min specs with specified VRAM, is that an estimated minimum based on fidelity of lowest settings?  Essentially, at min specs the frame will take X size, thus you need Y amount of vRAM??

Link to comment
Share on other sites

Link to post
Share on other sites

That is probably one of the best simple explanations I have read in a long time.

So with games having min specs with specified VRAM, is that an estimated minimum based on fidelity of lowest settings? Essentially, at min specs the frame will take X size, thus you need Y amount of vRAM??

Frame size only depends on the display, but the VRAM is used for some other things that I forgot about (got all caught up thinking about frame buffering, hehe) an example is textures. If you play Skyrim with a million billion texture packs I guess you could fill up an enormous amount of VRAM but memory bandwidth could still limit framerate. If the GPU has to wait in line to get all the texture information it needs to generate a frame, it won't be able to churn them out as quickly as it could otherwise. It's a tricky subject, even if you run monitoring software and see all 8GB of VRAM are filled, it doesn't mean memory bandwidth isn't limiting your framerate. You can be storing a lot of data but you might not be able to access it quickly enough to make 120 frames every second or whatever you would otherwise be capable of. So I guess it's possible to use it all, but like I said, can't use it effectively.

I would assume they get the minimum specs by playing the game at low settings at 720p and seeing how much it uses :D but yes in theory it would probably be a factor of textures and other data used by the GPU when everything is on lowest settings, estimate how much memory that would use, plus some additional capacity on top of that to be used for frame buffering at a normal resolution.

Link to comment
Share on other sites

Link to post
Share on other sites

Not really on topic, but whats the next generation after maxwell 970's ?

I think it's pascal.

The 256bit bus of the 970 probably wouldn't be able to make use of much more than 4GB.

Link to comment
Share on other sites

Link to post
Share on other sites

I think it's pascal.

The 256bit bus of the 970 probably wouldn't be able to make use of much more than 4GB.

Any release date on that?

planning to get a 970 this christmas.

dunno if i should buy one and risk a new pascal card coming out a month or two later.

Link to comment
Share on other sites

Link to post
Share on other sites

Any release date on that?

planning to get a 970 this christmas.

dunno if i should buy one and risk a new pascal card coming out a month or two later.

 

It's most likely going to be a while.  First gen Maxwell just came out last month and usually NVIDIA does 2 generations per architecture... so two years, if things go normally.

Link to comment
Share on other sites

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×