Written by Metal Messiah
It looks like we have some confirmation regarding the AMPERE GPU architecture from Nvidia. NVIDIA’s CEO” Jensen Huang” has just confirmed during a keynote press briefing that the upcoming Ampere architecture is going to launch for both “consumer” as well as the datacenter/HPC markets. This confirms all the previous rumors we have been hearing so far.
We will have Ampere lineup for the Gaming “Geforce cards” as well. The previous generation Volta and Turing series were actually “split” between professional and gaming segments. This changes with Ampere.
According to Jensen this will “streamline” the whole AMPERE GPU lineup. As posted by MarketWatch:
“” Ampere will eventually replace Nvidia’s Turing and Volta chips with a single platform that streamlines Nvidia’s GPU lineup, Huang said in a pre-briefing with media members Wednesday. While consumers largely know Nvidia for its videogame hardware, the first launches with Ampere are aimed at AI needs in the cloud and for research.
“Unquestionably, it’s the first time that we’ve unified the acceleration workload of the entire data center into one single platform,” Huang said. “”
Both the “Tesla” and “GeForce” Ampere lineups are going to have different configurations. TESLA will likely focus on compute applications (FP8/FP16 and FP64), while Geforce gaming series can benefit from faster FP16/32 cores as well. Jensen has confirmed that the GPU architecture will remain the same.
“Nvidia did not release any information about consumer GPUs using Ampere, but when asked by a reporter in the briefing about the difference between enterprise and consumer approaches to Ampere, Huang said “there’s great overlap in the architecture, but not in the configuration.”
User’s articles is a column dedicated to the readers of DSOGaming. Readers can submit their stories and the Editorial team of DSOGaming can decide which story it will publish. All credits of these stories go to the writers that are mentioned at the beginning of each story. Contact: Email
so….when? it its not out by the time cyberpunk 2077 comes out then im just gonna buy whatever is the newest amd. nvidia will have lost me forever.
“Forever”.
haha drama queen
I am the only one that thinks Cyberpunk graphics looks average. Hope they release a new game trailer so that we can check again!!
said the same thing a while back,… was criticised for it.
I agree we need to see more but given the state of their engine and the current console hardware I don’t think it will be anything more than average. I’ll be more than happy to be proven wrong though but CDproject’s showings don’t exactly leave any room for optimism
Hopefully even with those average graphics it runs at least at 60 FPS 1440p on 2070 Super but somehow I doubt it.
That’ll show ’em!
And at the low asking price of an healthy kidney 🙂
I read in news recently that during patient COVID examination doctors detected third kidney in him, so You know … lucky b**tard I guess ? 🙂
Unless Huang was exaggerating the Amperes should be about 50% faster than the Turings. That would be a really nice boost in performance.
Pricing is my biggest concern. I expect they are going to be even more expensive that Turings and Turings are already priced pretty high.
As per some rumors Ampere will only feature RTX cards, and no GTX variants.
This means all the cards are going to have RT and TENSOR cores on the hardware level, so this will make the DIE bigger in size, which will account for more production costs.
So I think these cards are going to be priced in same price bracket as Turing, unless of course AMD releases the rumored “Nvidia-killer” BIG NAVI RDNA2 GPU, which seems unlikely though, in my opinion.
AS much as I want NV to drop prices on Ti I doubt that this happen also.
This is my take on this:
AMD did announce PRO level GPU based on current gen arch – if they had something faster cooking up then they wouldn’t put that on professional level GPU – otherwise they would cannibalise their own product (consumer GPU faster then PRO offering).
Intel stated that they will not show info on theirs top end GPUs so my take on this is that this GPU will be still slower than Ti / Titan level GPUs from NV and IMO they plan to compete on price vs performance then on absolute performance field.
NV moves all to RTX so they don’t need dual design so it’s, IMO, cost saving appraoch (no need to additional dev for GTX side of things anymore).
I can’t recall in recent history price addiustment for new product line (at least not in down direction).
The thing with 1080Ti was VEGA overhype and because AMD did released that PRO GPU I doubt that this will happen with RDNA2 for them so NV don’t need to compete this release round.
Note that 2080Ti is twice as fast as 1080Ti but in rendering / Ai fields. For gamers totally useless but for 3d artists this is actually very handy – much faster previews of work. And as usual I think NV will market this as:
– want to game on “cheap” go console
– want gaming PC go up to xx80 level GPUs
– want to do work with 3D go xx80Ti / Titan level
etc.
Yeah, Intel will mostly target the Mid-range and “mainstream” GPU market segment, at least when it comes to Gaming.
Don’t expect a high-end performing GPU like the RTX 2000 series, or the RX 5000 XT cards from AMD.
Raja has made this very clear that Intel is only going to release “budget” GPUs for gamers, so that everyone can afford them.
He will sell his “Leather Jacket” as well.
This guy is damn RICH though.
Nvidia has certainly made a lot of money for Huang. His wealth is close to 7.5 billion dollars. He wasn’t always wealthy though. In the early 90s Huang and other co-founders met in a Dennys and decided to start Nividia.
Wow !
Indeed. But why can’t NVIDIA reduce the price of their GPUs and make it more competitive and “mainstream” ? They have already dominated the high-end GPU market segment.
But I know some gamers are willing to pay a premium, so why will the company bother for the average gamer !
Blame AMD lack of high end.
Hopefully that changes this year though, Competition is great and look at the shake up in cpu’s in the past 2 years, Complete reversal.
Yeah lol, I’ve been constantly blaming AMD for not giving us an RTX Ti killer !!
I know because of AMD, nvidia is not forced to reduce the price of their cards, but still the company which has earned billions of dollars can consider an average gamer’s woe.
On the CPU side, yes, it’s the totally opposite. AMD has been quite dominant with the release of Ryzen, and they have shaken the whole CPU market by storm ! This has forced INTEL to re-think about its pricing policy, and Intel has also been relying on the same old aging process node, with no other option than to release CPUs with a higher clock speeds, resulting in insane power consumption as well.
IMO, AMD has totally dominate the whole CPU landscape. They have also eaten up the market share from INTEL. Ryzen CPUs are far more efficient than INTEL.
they were the only player on the high end right now. so the performance come with big premium. but in mid range and below (RX5700 is more like high end to me than mid range) nvidia still retain their competitiveness in pricing and it really makes AMD heads spinning to counter them. yes AMD often touted as value/budget champion but that’s only from consumer perspective. for AMD being that kind of champion did not get them much profit (and most of their sales coming from this segment) and despite having that crown for almost 10 years their market share (for discrete GPU) still lingers around 30% vs nvidia.
when AMD priced their RX590 for $275 in late 2018 it seems like an indirect signal to nvidia that they want polaris level of performance to stay on $250 price mark rather than making it more cheaper after two years (after seeing how nvidia price their turing). but nvidia still carry out their attack anyway with 1660Ti. then we saw RX5500 did not do the usual undercutting that they always do with nvidia and finally the mess we saw around RX5600 Bios. the mess happen because AMD can’t further reduce their price than they already did when nvidia simply drop RTX2060 price down to $300.
so imagine if nvidia retain similar price structure as pascal with turing.
$700 RTX2080Ti
$550 RTX2080
$400 RTX2070
at this time AMD fastest GPU is Vega 64 which usually sold around $500 brand new. it has to compete with faster and yet cheaper 2070? the bomb will be when nvidia release their RTX2060 for $250 in early 2019 that have similar performance to AMD Vega 64. at this point RX5700 still at least 5 months away. if nvidia really do this kind of pricing we will never ever see Radeon 7. Vega 64 will be forced to sell under $250.
the most ironic thing about nvidia price increase that many people did not like is they give AMD some room to breath before they start launching their Navi. but even with that what nvidia has done on the mid range market still enough to give AMD the headache because of the very low profit margin on this segment. that’s why for years i have been saying that nvidia is more ruthless than intel.
I agree with most of your points…..Very well put !!