Friday, March 11th 2016
NVIDIA "GP104" Silicon to Feature GDDR5X Memory Interface
It looks like NVIDIA's next GPU architecture launch will play out much like its previous two generations - launching the second biggest chip first, as a well-priced "enthusiast" SKU that outperforms the previous-generation enthusiast product, and launching the biggest chip later, as the high-end enthusiast product. The second-biggest chip based on NVIDIA's upcoming "Pascal" architecture, the "GP104," which could let NVIDIA win crucial $550 and $350 price-points, will be a lean machine. NVIDIA will design the chip to keep manufacturing costs low enough to score big in price-performance, and a potential price-war with AMD.
As part of its efforts to keep GP104 as cost-effective as possible, NVIDIA could give exotic new tech such as HBM2 memory a skip, and go with GDDR5X. Implementing GDDR5X could be straightforward and cost-effective for NVIDIA, given that it's implemented the nearly-identical GDDR5 standard on three previous generations. The new standard will double densities, and one could expect NVIDIA to build its GP104-based products with 8 GB of standard memory amounts. GDDR5X breathed a new lease of life to GDDR5, which had seen its clock speeds plateau around 7 Gbps/pin. The new standard could come in speeds of up to 10 Gbps at first, and eventually 12 Gbps and 14 Gbps. NVIDIA could reserve HBM2 for its biggest "Pascal" chip, on which it could launch its next TITAN product.
The GP104 will be built on the 16 nm FinFET process, by TSMC. NVIDIA is hoping to unveil the first GP104-based products by April, at the Graphics Technology Conference (GTC) event, which it hosts annually; with possible market availability by late-May or early-June, 2016.
Source:
Benchlife.info
As part of its efforts to keep GP104 as cost-effective as possible, NVIDIA could give exotic new tech such as HBM2 memory a skip, and go with GDDR5X. Implementing GDDR5X could be straightforward and cost-effective for NVIDIA, given that it's implemented the nearly-identical GDDR5 standard on three previous generations. The new standard will double densities, and one could expect NVIDIA to build its GP104-based products with 8 GB of standard memory amounts. GDDR5X breathed a new lease of life to GDDR5, which had seen its clock speeds plateau around 7 Gbps/pin. The new standard could come in speeds of up to 10 Gbps at first, and eventually 12 Gbps and 14 Gbps. NVIDIA could reserve HBM2 for its biggest "Pascal" chip, on which it could launch its next TITAN product.
The GP104 will be built on the 16 nm FinFET process, by TSMC. NVIDIA is hoping to unveil the first GP104-based products by April, at the Graphics Technology Conference (GTC) event, which it hosts annually; with possible market availability by late-May or early-June, 2016.
135 Comments on NVIDIA "GP104" Silicon to Feature GDDR5X Memory Interface
It is all speculation 5X or not, the real benchmarks on real production silicon will show how really works. HBM yes or not doesn't really matter as long it works. And I haven't seen any info on bus width for these products too.
@Soalris17 I don't think 980ti is still and will be a fine card... anything more powerful justifying an upgrade cost still won't be really seen till late Q3, and that's hell of a lot time actually, so enjoy it...
As buyers we still will wait till both camps release Polaris and Pascal and then decide what is the best choice.
and why people talk about asyc, none game support that for now, and just small performance for FPS or TPS game
Gratz, corporate BS actually got through your thick skull and made you believe this stuff.
Ask your GTX 980 Ti using friends how miserable they are that they don't have the HBM that the Fury X uses.
Any thoughts on if new motherboards will need to be manufactured to support the new gpu technologies?
Probably that's a hidden announcement that says: "let GFX warz startorz" on this site.
Startorz they have.No, certainly not.
As for the Anandtech article - it's a bit meh.... And I've lost alot of faith in Appletech, I mean, Anandtech.
you _cant_ trust that shit, especially given nfooledya track record
they _will_ use gddr5x for mid/low end and thats fine, hbm doesnt even make sense here since the gpu doesnt have enough fillrate to saturate the bus (red camp will do it for sure too). this has been known for quite a while yet people like to speculate this and that
However - if you only run a single GPU then I advise you to stop living in the past.
This thread is also an Nvidia thread btw so please quit with the AMD bashing as its not relevant here
Yet again, this is the internet - trolls like to throw anything around.
I would consider getting an AMD GPU if they actually offered a superior product to the 980 ti. I actually came here looking for feedback and suggestions. Does AMD have a product that would make more sense than going with the 980 ti?
With that truly we will move to the UHD
Just my two cents
@HumanSmoke does know what I was thinking, because unlike you, he actually understands the concept of context, and has had frequent interaction with me, both positive and negative.
If you don't understand my point that anger over a GPU company is nonsensical, because it does not make any difference in your life like real life events do, then I'm sorry. I can't create rational thought in your head. It doesn't matter if this is a tech site. Over the top anger at a hardware company is completely irrational, as they miniscule in terms of their impact on life.
I just noticed how young you are. This explains your non-comprehension of what's actually important in the world. Like I said, you've got the luxury to allow yourself to get worked up about GPU's. It still can't ruin your life though. One day you'll realize what is actually worth getting mad over.
I'm done with this thread.