Monday, September 17th 2018
NVIDIA Segregates Turing GPUs; Factory Overclocking Forbidden on the Cheaper Variant
While working on GPU-Z support for NVIDIA's RTX 20-series graphics cards, we noticed something curious. Each GPU model has not one, but two device IDs assigned to it. A device ID is a unique identification that tells Windows which specific device is installed, so it can select and load the relevant driver software. It also tells the driver, which commands to send to the chip, as they vary between generations. Last but not least, the device ID can be used to enable or lock certain features, for example in the professional space. Two device IDs per GPU is very unusual. For example, all GTX 1080 Ti cards, whether reference or custom design, are marked as 1B06. Titan Xp on the other hand, which uses the same physical GPU, is marked as 1B02. NVIDIA has always used just one ID per SKU, no matter if custom-design, reference or Founders Edition.
We reached out to industry sources and confirmed that for Turing, NVIDIA is creating two device IDs per GPU to correspond to two different ASIC codes per GPU model (for example, TU102-300 and TU102-300-A for the RTX 2080 Ti). The Turing -300 variant is designated to be used on cards targeting the MSRP price point, while the 300-A variant is for use on custom-design, overclocked cards. Both are the same physical chip, just separated by binning, and pricing, which means NVIDIA pretests all GPUs and sorts them by properties such as overclocking potential, power efficiency, etc.
When a board partner uses a -300 Turing GPU variant, factory overclocking is forbidden. Only the more expensive -30-A variants are meant for this scenario. Both can be overclocked manually though, by the user, but it's likely that the overclocking potential on the lower bin won't be as high as on the higher rated chips. Separate device IDs could also prevent consumers from buying the cheapest card, with reference clocks, and flashing it with the BIOS from a faster factory-overclocked variant of that card (think buying an MSI Gaming card and flashing it with the BIOS of Gaming X).
All Founders Edition and custom designs that we could look at so far, use the same -300-A GPU variant, which means the device ID is not used to separate Founders Edition from custom design cards.
We reached out to industry sources and confirmed that for Turing, NVIDIA is creating two device IDs per GPU to correspond to two different ASIC codes per GPU model (for example, TU102-300 and TU102-300-A for the RTX 2080 Ti). The Turing -300 variant is designated to be used on cards targeting the MSRP price point, while the 300-A variant is for use on custom-design, overclocked cards. Both are the same physical chip, just separated by binning, and pricing, which means NVIDIA pretests all GPUs and sorts them by properties such as overclocking potential, power efficiency, etc.
All Founders Edition and custom designs that we could look at so far, use the same -300-A GPU variant, which means the device ID is not used to separate Founders Edition from custom design cards.
90 Comments on NVIDIA Segregates Turing GPUs; Factory Overclocking Forbidden on the Cheaper Variant
That would be like saying the 1080 Ti is almost the same as the GTX 970 since they have similar die sizes - except you can't say that because you are comparing 16nm dies to 28nm dies. They are different processes with different capabilities.
Fact:
-100
-102
-104 Midrange
-106 Low End
-108
No thanks.
GX-102/100: Enthusiast
GX-104: High end
GX-106: Mid range
GX-107: Low end
GX-108: Entry/mediaWhere exactly? He said it's have been very restricted, but not exactly locked altogether.
Steve: Community decision is that its locked, Tell me why i'm wrong
Tom: Your not wrong
You can still manually overclock all GPU's the only thing is now the chips are presorted according to their capacity (something like this is already happening in RAM samsung vs micron chips...). IF this clarity is passed on to consumer i am all for it.
Now you can in real time choose based on your preference, the only caveat being device ID are hidden or rather not shown on the box so how will we know which is which?
Modern vehicles have more driver aids than ever before, and younger drivers seem to know less and less about what makes them tick, or how to repair anything, let alone change a flat tire.
With self driving cars being seen as the future (and Nvidia involved heavily in that) drivers wont need to know anything other than how to get in and set the destination.
AMD really does face an uphill challenge when you have sheep throwing their own koolaid at other people. Also you do notice your own twisted marketing defense still places the $600 "2070" as mid range... right? That's fine, go buy Nvidia's newest $600 550 Ti if you want to.... smh
having said that, i managed to pre order a gigabyte 2080 Ti "windforce OC" card for $1030 after tax. given that it is a factory overclocked card, i guess it means that it's a binned GPU? that would be awesome.
In fact, for some of us, it might actually be a good thing? What if you could buy the bullshit model and flash it with the 1337 BIOS? BIOS editing days are pretty dead, at least for nVidia, but at least you can try flashing a premium bios on a bland model.
no difference to the enduser
The only problem then is the same old price issue. ie we don't like the trend of increasing card prices. Competition is the only thing that will fix that. Of course that means AMD needs to take note of how much NV is able to make off each card. Even if PC's aren't really their focus right now, business models can shift if there's money to be made.
After this bit of news dies down, next generation I would imagine nvidia will again tighten the noose more to wring more money from their client base while at the same time assuming more control over what their AIBs can& can't do, if they still want them at that point.
- Voltage: 1.09V
- Power: TDP+20% (somewhat configurable by manufacturer, I have seen 13% to 30%)