Rumor: Nvidia GeForce 1180 / 2080

[ShogoXT]

There has been information floating around that the next gen of GeForce cards will be introduced September.

First will be the the xx70 and xx80, with the other tiers later. GeForce RTX branding has also been trademarked. Code names have actually been unclear as Ampere and Turing names have come up. Volta has been mentioned as well, but Volta is likely limited to workstation and compute.

Major new features include use of GDDR6 and Ray tracing.

Nvenc hasn't been mentioned and it's likely too early for any AV1 support. Vp9 improvements would be nice.

[ShogoXT]

https://videocardz.com/77498/nvidia-announces-geforce-rtx-2080-ti-rtx-2080-and-rtx-2070

RTX 2000 Series Announced

MSRP Prices:

RTX 2080 Ti 11GB GDDR6 $999

RTX 2080 8GB GDDR6 $699

RTX 2070 8GB GDDR6 $499

Founders Edition is an additional $100 each, with TI being $200 more.

The event was mostly about ray tracing and real time surface reflections. 8k decoding was rumored for content creators, otherwise I didnt see anything mentioned about Nvenc unfortunately.

[nevcairiel]

Can probably expect encoding improvements, as those have been rumored before. Decoding, there really isn't much to do right now. AV1 would be nice, but its probably too early.

Interesting would be if real-time upscaling using the Tensor Cores with Neural Networks would be something that can be used much more performant in video renderers, now that the Tensor Cores are coming to all NVIDIA cards, and not only the workstation cards.

[Gser]

Wow that's just ridiculously expensive. Guess I'll be sticking with AMD from now on.

[huhn]

the 2070 is supposed to outperform a titan Xp so the prices are "fine"

if these are binned turing quadro cards than the 2080 ti is 754nm that's pretty big...

but these tensor core are a headache to me. dedicated hardware for something like physx. all console are using AMD cards if AMD isn't adding something like this too this tech will die just like physx. and coding something only for RTX card sounds like a waist looking at the limited user base.

i guess i will skip them too was really looking forward to them but they are nearly for sure 12 nm and 7 nm cards are not that far away if consumer or not well i guess i have to wait and see.

[NikosD]

Can probably expect encoding improvements, as those have been rumored before. Decoding, there really isn't much to do right now. AV1 would be nice, but its probably too early.


I could hope for a 2050 (Ti) late announcement around November, suitable for AV1 implementation.
But again it's only a three months difference.


Interesting would be if real-time upscaling using the Tensor Cores with Neural Networks would be something that can be used much more performant in video renderers, now that the Tensor Cores are coming to all NVIDIA cards, and not only the workstation cards.
Technically, Volta Tensor Cores have already been released to a prosumer card, Titan V, but definitely a very high priced card around 3500$ and very powerful too.

I'm really curious if 2060 and 2050 (Ti) will have Tensor cores and Ray Tracing cores inside.

the 2070 is supposed to outperform a titan Xp so the prices are "fine"

Hahahahaha.

You are so funny guy!

Do you really believe the "leather-jacket-man" that he was referring to rasterized gaming performance ?

He was referring to some strange benchmark comparing Tensor or Ray Trace core performance mixed with CUDA cores performance.

Every other comparison is ridiculous.

[huhn]

i said supposed but the RTX 2080 has a TDP of 250-285 which is more or the same as the titan Xp has with a better process so.

[nevcairiel]

but these tensor core are a headache to me. dedicated hardware for something like physx. all console are using AMD cards if AMD isn't adding something like this too this tech will die just like physx. and coding something only for RTX card sounds like a waist looking at the limited user base.

AI is here to stay. Its seeing much larger adoption everywhere, and the usecases are endless, and certainly useful in gaming. The entire industry is jumping onto this.

The same could be said for the Ray Tracing cores. Ray Tracing is the holy grail of 3D rendering. And a major engine like UE4 adding support immediately for everyone to use will make it relevant, plus EA's Frostbite engine (BF5 etc) and a bunch of other game studios. Those are engines that are also used with console games - the non-Ray Tracing rendering isn't going away, it just won't look as pretty.

AMD really doesn't have a choice but to do something about either of those. These features are too huge to ignore and hope they go away.
AI might be an "addon" for gaming, but there will be loads of workstation tasks that will make use of those, and AMD has been trying to get into that field. Ray Tracing, I believe, will really make game developers happy, because it makes things just so much nicer.

[huhn]

are they to big to be ignored? the next consoles are going to be vega APUs do you really see them adding to these type of cores to the GPUs. do they even have the time

i'm a PC gamer myself but i know very well that we PC gamer are not that important sad but true.

if nvidia will release GPU without it and that is highly possible it will have a hard time.

even there "demos" with on/off look not that great... https://twitter.com/NVIDIAGeForce/status/1031600171241463808

these engines can do physx too and this makes at least a huge difference unlike ray tracing. ray tracing is pretty much godrays in fallout 4 if you like the nvidia mod vault 1080 well yeah looks a lot different in the game but in the rest of the game well people can live totally fine without.

it currently even reminds me of stuff like motion blur or depth of field so stuff that makes it look more realist but harder to see what is happening and they are not that popular or let'S say not seen as useful to everyone.
and no it's not use useless but physx is not useless too but who cares about it and that'S my point?

a couple years back VR was the holy grail of 3D rendering where is it now?

and for workstation we have quadro cards.

[nevcairiel]

a couple years back VR was the holy grail of 3D rendering where is it now?


Ray Tracing has always been the Holy Grail of 3D rendering since its inception almost half a century ago, it just wasn't achievable in real-time.
Any movie-style scene that was rendered off-line is ray traced, because its the only way we currently have to make something really look good.

VR really isn't that interesting from a rendering-side, nothing much changes for the engine itself, it just has a different FoV and renders two eyes, but same concepts all around.

In a couple years, when Ray Tracing becomes the standard of everything, game developers will sigh in relief as they can stop with the ugly hacks and cheats to make scenes look acceptable. Thats the real beauty of ray tracing. You just build the scene and it automatically looks accurate. Right now you have to manually fix every reflection, every shadow.

[huhn]

and fallout 4 has "ray tracing" in real time it's not 100 % the same but comes close. not a pretty game but what ever.

it's said that RTX is using a part of DX12 with is good but it is on the other hand part of nvidia gameworks so AMD can not use the "same" ray tracing so this is already half a nightmare even through AMD has it own ray tracing.

if it would be something like x86 in CPU and AMD could use the same thingy and create there own cores to the the same type of ray tracing i wouldn't see a big problem here but as it looks now AMD has to do something else to do the same ray tracing and that'S a problem the results will not be the same. so you game developer have to create 2 or even 3 types of ray tracing.

and if we come now back to video rendering using NN operation oyu currently have to create one precisely for nvidia and only for nvidia because AMD/ intel will not output the same number if you ask them to do the same.

[NikosD]

Nvidia is a highly aggressive and highly unethical company, take it as a fact.

The only thing that it really cares is to destroy the competition.

There is only one company left, AMD.
All the others are gone because of Nvidia.

GeForce Partner Program (GPP) was the last attempt to destroy completely AMD but fortunately, it didn't work.

GameWorks, HairWorks, PhysX are all parts of his ambition to eclipse the competition and to be alone, a great monopoly in the GPU arena.

Now, with RTX cards, Nvidia is trying to sell in HUGE prices things like Tensor cores which are completely useless in general for gaming and Ray Tracing cores which are useless right now, as no game is actually using this technology.

Gaming for Nvidia is no longer a priority, although it keeps bringing most money of all other categories together.

Tensor cores are only used for AI and Deep Learning/ Machine Learning and are actually used by Quadros and Teslas.

Ray Tracing could be useful in the future, but not right now.

But Nvidia asks for your money right now, while the rasterized gaming performance of the new cards is certainly not justifying their price.

Do you have an arm or a leg more ?

Sell it and buy a 2080 Ti for 1500€.

P.S

Don't be surprised if you find out that new games coming do not work properly in older than Turing generation of Nvidia cards or AMD cards.

It's part of the plan.

[Atak_Snajpera]

If you like pathtracing in games then you should try quake2 (do not forget to set 640x480 resolution for full 90s experience)
http://amietia.com/q2pt.html

[Rumbah]

Any movie-style scene that was rendered off-line is ray traced, because its the only way we currently have to make something really look good.
Well, I guess it's not that easy.
For example Cars was Pixar's first completely raytraced movie if I remember correctly. Everything before was at least hybrid. And even today complete raytracing is slow and movies use something like the unreal engine for their effects (of course it may also depend on the budget).

And it can be harder to use for an artist. If you want the clean realistic look it's fine. But if you want to change the look or make adjustments to the lighting of the scene you have to work with all the (diffuse) reflections of your light sources. That can make it much harder to get the lighting you want artistically.

[NikosD]

RTX of Nvidia has nothing to do with pure ray-tracing.

You need Petaflops and not Teraflops for this.

It's a hybrid and very slow implementation of ray-tracing mixing everything (software, SDK, Tensor cores, Ray Tracing cores and shading CUDA Cores)

This generation of RTX implementation, Turing cards, should be skipped by gaming users as even the fastest card of 1400€ - 1500€ - the 2080 Ti - throws back the resolution and FPS to 1080@30fps when RTX is on.

When more and optimized games for Ray-Tracing show up, if ever, this generation of cards will be obsolete.

I think the next architecture of Nvidia at 7nm is not far away - probably less than a year - and will be more suitable for Ray-Tracing, if there is a market for such thing in gaming.

[Asmodian]

It should be good for Tensor Flow hobbyists. :D

[huhn]

the first benchmark leaks are spreading.

the 2080 looks like it will be a faster than a 1080ti which means the cuda cores got quite a lot faster. so it looks like ~0-30% more performance per dollar in classic application. i guess it is really important how the prices are going to be a week or two other the release.

[NikosD]

I take the risk to say that 2080 will be around 8% faster than 1080 Ti and probably will loose in specific game settings and resolutions.

[Blue_MiSfit]

I'm sure its performance in traditional tasks will be a nice upgrade.

Keeping my eyes out for the dream display - a ~27-30" 4k @ 144 Hz HDR IPS LCD display with 1000+ nits and G-Sync. Pair that up with a new 2080 and I'll be happy as a clam for the next couple years :D

[huhn]

it is not really in a question if these cards will be faster the 2080 ti will be faster than the 1080 ti the real issue is the price.

if a 1080 ti cost as much as 2080 and has the same performance than this is an issue still favouring the 2080 but still that's a weak showing.

but i will do my judgement on this about 2 week after the release and see what the prices are then.

but i have to say i can't see high setting 60 fps 4K gaming with these cards not even the 2080 ti. newer games are not going to be easier.

but here is your screen: https://www.acer.com/ac/de/DE/content/predator-series/predatorx27

[NikosD]

We never compare models but prices.

2080 Ti has a 70% initial price more than 1080 Ti.

Is it going to be 70% faster than 1080 Ti in traditional games ?

Certainly no.

2080 has a price closer to 1080 Ti, that's why I did the comparison between them.

Since Ray-Tracing in gaming is simply not existent and with an unknown future, I can't find a reason to buy RTX cards at these prices.

They simply don't worth their price.

[huhn]

if the planned US none founder edition MSRP are real(1000 and 700 respectively) it's not to bad paying a premium for a flag ship is nothing new.

the 2080 ti has about 48 % more cuda cores than a 2080 and cost about 50 % more but that doesn't change the performance per dollar figures.

[NikosD]

Just announced MSI prices here in Greece.

2080 Ti ~ 1600 €

2080 ~ 1100 €

Is nVidia a completely insane company ?

Are the clients of nVidia completely insane ?

Will see...

[huhn]

and the taxes in greek have nothing to do with that?

and can you link the announcement plz.

and the more you look into these cards the more sense the prices makes and the less sense this whole product makes to me.

the 2080 ti is a ~750 mm² die that's 60 % more than a 1080 ti. this alone makes the card every expensive.

[NikosD]

The VAT here is 24%

The cards are very slow on RTX games and rushed to the market in order to hit AMD and differentiate further nVidia from AMD.

They are completely meaningless right now and with these prices nVidia should offer shares of the company, because it's like that.

It's like you are going to participate in their R&D expenses for something with no use and uncertain future.

We, as customers, don't care if companies are arrogant and make mistakes.

We are not going to pay their unreasonable ambitions.

The more I think about Turing of nVidia the more it looks like 10nm of Intel.

A disaster for the arrogant companies.

https://www.plaisio.gr/search.aspx?query=2080&catalogue=all&mode=searchlist

[huhn]

that's not an announcement from anyone.
or is this one too: https://geizhals.de/?cat=gra16_512&xf=9810_9+11750+-+RTX+2080+Ti

are they slow? they make bigger cards with relative performance taking the size into account and adding some goodies you may never use at all no one forces you to use ray tracing and you can do other things with these tensor cores to which is supposed to be giving you more performance but no one knows how this new fast AA looks.

if this is bad what is vega?

RX 580 is 230 nm and VEGA 64 is 510 nm and only 50 % faster.

don't get me wrong i'm not thrilled that the end user get's the same performance per dollar 2 years later this looks like a "refresh" with much bigger cards.

[NikosD]

What you say, just doesn't make sense.

You don't play games by the die size.

You don't buy cards by the die size.

You are not an investor to support companies.
Period.

[huhn]

nvidia has to pay for the die size...

a GPU with a bigger die cost more money in production. you are not going to pay more for production and ask less money for it that doesn't make sense.

[NikosD]

Oh my...

Things are even worse than I thought they were.

Nvidia’s Director of Technical Marketing, Tom Peterson, joined HotHardware on its 2.5 Geeks podcast and when queried about the RTX 2080 outperforming the GTX 1080 Ti, he said that he thinks there would be cases that would happen but couldn’t say for sure.

So, in all previous releases the xx70 outperformed the xx80 Ti card.

For Turing release, the Director of Technical Marketing of Nvidia, is NOT SURE if even 2080 card can outperform 1080 Ti.

What a disastrous performance!

What a disastrous release!

[nevcairiel]

So, in all previous releases the xx70 outperformed the xx80 Ti card.

That statement isn't really accurate. It was only true for Pascal (ie. a 1070 vs 980 Ti). One generation back, a 970 does in fact not beat a 780 Ti. And before that, a 680 Ti didn't exist. A 690 did, but that was a dual-GPU card, so thats quite a different beast. One generation does not make for an "all previous releases".
The conventional wisdom before Pascal used to be that with a generational change, the models move one up in the performance scale - not two. Pascal was the exception.

[huhn]

the big problem here is 12nm is not a real shrink compared to 16mn it's mostly more power efficient and doesn't allow better clocks.

[NikosD]

That statement isn't really accurate.
The conventional wisdom before Pascal used to be that with a generational change, the models move one up in the performance scale - not two. Pascal was the exception.

970 was very close to 780 Ti and it depends on the clock (overclocking) to find the winner.

1070 was clearly faster than 980 Ti.

So, the new norm after 2 years and 3 months of the latest release could someone expect to be the latest one - the Pascal's model.

Pascal could be considered not as an exception, but as the new norm since it's already too old and we only have two generation cards to refer to (970 vs 780 Ti and 1070 vs 980 Ti)

But even using your standards:

2070 is not close to performance to 1080 Ti.

It's a lot slower, could be like 50% slower, it's not close like 970 vs 780 Ti

I doubt 2070 will be faster than 1080, the previous model.
It will struggle to defeat it, if ever.

But we actually have a problem here.

Because your beloved friend, Mr. Leather Jacket Man, said than 2070 is faster than Titan Xp!

So, I actually agreed with him expecting 2070 to be faster than 1080 Ti.

And our friend here, huhn, believed him!

He believed the crook.

And now believe me the phrase "2070 is faster than Titan Xp" is a lot more inaccurate than "xx70 models were faster than xx80 Ti models before Turing cards"

But I am with you.

So far it seems that 2070 will be struggling to be faster than 1080 the previous generation one model up.
(we don't even dare to compare it with 1080 Ti)

2080 is not faster than 1080 Ti.

According to Nvidia it struggles to go ahead of the previous model.

Even with your standards, by all means, the Turing RTX 2000 cards is the definition of performance regression after more than two years of the latest release and the definition of overpriced cards.

Extremely overpriced cards.

They simply don't deliver.

[huhn]

so saying "the 2070 is supposed to outperform a titan Xp" is believing?

and you are ignoring that maxwell is 28 nm and pascal is 16 nm.
this is a very important difference and the reason i think Turing should be delayed for 7 nm. as far as i "know" 7 nm is ready for mass production.
and again 12 nm is just a 16 nm with an intel + so they use the "same" node there is not much you can gain from doing that.

the 2070 will have about 75 % of the performance of a 2080 make it a card that will be close to a 1080.

[NikosD]

So, we all agree that nVidia made a huge mistake by releasing this premature architecture so early, so overpriced, so damaged in performance.

And we should warn potential customers to stay away from this generation of cards - Turing cards.

Clients shouldn't pay the mistake of nVidia.

Agreed.

[nevcairiel]

Everyone should always judge hardware based on actual performance numbers, as provided by independent reviewers. We don't have those yet.

If those rumored numbers come close to reality, it may not be an upgrade for people that already own a 10-series card, but if you don't, might as well get a 2070 instead of a 1080 in the future (once prices settle weeks/months after launch). Unless of course you want a 2080 Ti with its rumored 40-50% increased performance over a 1080 Ti, afterall its the only way to get that level of performance, even if it costs you. But people also bought Titans before, which cost the same, and this time there is actually a huge performance difference, contrary to the Titans which were only a small upgrade over the Ti's of their time.

[NikosD]

But the problem is that 2070 is going to be slower than 1080, by a good chance.

Slower and with higher price, for so long time that probably a new architecture will come because there is no 2070 yet and in the next month.

Also for 2080 Ti latest rumors place the card around 25% faster than 1080 Ti with double the price!

It's simply not worth it.

[huhn]

you can rant about the prices 2 weeks after the release.
the planned MSRP for the 2070 is 500 dollars.

on paper the 2080 ti has 50 % more power than a 2080 and the price is about 50 % more.

the number of people buying a slower card for a higher price is not very high then they will simply not sell.

the vega 64 is kind of an alternate product to the 2070 so nvidia can't dictated the price of the 2070.

[Blue_MiSfit]

Not bright enough :) I'll wait until I can get one with 1000 nits. That one is only good for 400 average IIRC

[huhn]

it can do the vesa HDR 1000 the highest specified HDR spec i'm aware of.
you can get brightness numbers here:
https://www.rtings.com/monitor/reviews/acer/predator-x27

even with 10% it is far over 1000 nits. totally able to make you nightblind for the next 5 mins.
with an IPS typical "astounding" CR of 800-1350

[huhn]

the rumor about no tensor cores for 2060 or just staying with pascal are getting more clear.

while i can totally understand that a slower GPU will not be able to do ray tracing i have a hard time understanding why they should not be able to do DLSS or other potential use cases for tensor cores.
when i look at the market share of high end GPUs like 1080 or 1080 ti i you can clearly see a share about 4 after years of availability i see a problem in reaching the same number with msrp of the 20xx series in just a couple of years.

or with other word i don't see more than 2-6 % of user having tensor cores in the next 2 years. if the main stream cards don't have tensor cores.

i "predict" these numbers based on steam surveys: https://store.steampowered.com/hwsurvey/Steam-Hardware-Software-Survey-Welcome-to-Steam

[NikosD]

the rumor about no tensor cores for 2060 or just staying with pascal are getting more clear.

while i can totally understand that a slower GPU will not be able to do ray tracing i have a hard time understanding why they should not be able to do DLSS or other potential use cases for tensor cores.

You can't actually predict the behavior of such a greedy and ruthless company like nVidia.

But if 2080 or even 2080 Ti have so many performance problems with RT, I don't expect an RTX 2060 or lower card.

I'm not even sure that will see any 2060 or lower Turing cards (GTX or RTX) before Pascal sells every stock or ever.

Regarding DLSS, we don't know yet how much is taxing Tensor cores and if the general architecture of Turing is meaningful if everything is ON (RT cores, AI cores, CUDA cores)

[videoh]

such a greedy and ruthless company like nVidia It's called free enterprise, aka capitalism, buddy. Grow up.

[NikosD]

It's called free enterprise, aka capitalism, buddy. Grow up.Hahahahaha.

There are a lot of differences in the implementation of capitalism.

AMD and Xiaomi for example are still companies seeking for profit but as the CEO of Xiaomi has already said too many times, their profit is around 8% and they are going to push it down to around 5%.

So, nVidia by eliminating competition all these years with unethical and actually illegal means, wants to enforce not some kind of rules of the free enterprise but rules of the jungle.

They want to be a covered monopoly in order to milk everyone as much as they can with no rules and no other choices (competition)

Intel and nVidia are among the most hated tech companies of the world for a reason.

And please, don't be a blind nVidia fanboy.

Grow up.

[videoh]

A Greek is lecturing us on economics. LOL!

[Gser]

Allow me to rain on both of your parades. Xiaomi is limiting short turn profits to starve out competition and grab as many customers into its app/cloud eco system which it sees as the real money maker (this comes directly from a Xiaomi press release). All the big companies that form the oligarchs have to resort to unethical business practices e.g. treat their employees inhumanely, destroy the environment by cheaply sourcing materials and manufacturing obsolescence, bribing politicians for favorable laws and kickback investments, resort to false marketing, to survive in our late stage capitalist world that is run entirely by the stock markets.

[NikosD]

A Greek is lecturing us on economics. LOL!Interesting...

So, besides a blind nVidia fanboy and fan of boundless capitalism you are a racist too.

I'm on Tapatalk right now and I can't see if you have any info regarding your country.

Where are you from ?
Germany ?
LOL!

[NikosD]

Allow me to rain on both of your parades. Xiaomi is limiting short turn profits to starve out competition and grab as many customers into its app/cloud eco system which it sees as the real money maker (this comes directly from a Xiaomi press release).
No, Xiaomi can't eliminate the competition.

It's fourth after Samsung, Huawei and Apple and it will never reach them.

They give the best phones for their value , best VFM and they simply don't sell overpriced because they don't find a reason (!)

For all the other things that you said I agree with you, but what I'm trying to say is that not all companies are the same.

No AMD and Xiaomi are not nVidia and Intel.

You should probably study better the history of both companies to find out unbelievable facts about them.

[Gser]

No, Xiaomi can't eliminate the competition.

It's fourth after Samsung, Huawei and Apple and it will never reach them.

They give the best phones for their value , best VFM and they simply don't sell overpriced because they don't find a reason (!)

For all the other things that you said I agree with you, but what I'm trying to say is that not all companies are the same.

No AMD and Xiaomi are not nVidia and Intel.

You should probably study better the history of both companies to find out unbelievable facts about them.
It's not aiming for Samsung or Apple because it can't compete with their R&D and brand value. It's trying to flush out other manufacturers at the same price point like oneplus, Huawei and motorola. It won't be successful of course because it's phones are not as good as oneplus, and motorola has better value for money and its cloud services are complete shit compared to Google's that most Android users use. It also faces stiff competition and an entrenched market filled by Huawei, Oppo and Vivo in the asian market. Pocophone is a good effort, we shall see if it makes any dent in the market.

You are right they are not like nVidia or Intel because they can't afford to be like them because they lag so much behind in market share and sales and technology. AMD and chinese phone makers are known as product suppliers of the poor for a reason.

[NikosD]

It's not aiming for Samsung or Apple because it can't compete with their R&D ... AMD and chinese phone makers are known as product suppliers of the poor for a reason.

I have to disagree with you on most of your points.

Huawei is a huge company even bigger than Intel and is number two right now at smartphones just after Samsung.
Its aim is to surpass even Samsung in 2019, so it's out of the Xiaomi's targets.

AMD and Xiaomi are not making products for the poor.

They are making good products at sane prices for sane people.

EPYC CPUs and Threadripper CPUs are not cheap, but compared to Xeons they are cheaper for their performance and in sane prices for their value.

AMD has better CPUs now at all categories than Intel.
They are always faster on multithreaded apps and always cheaper.

The only thing missing from AMD is giving money to tech journalists in order to write reviews in a form of propaganda.

Like what Intel always did in the past and does even now with Zen architecture.

They are literally buying time to catch up with Zen architecture and 7nm of TSMC but of course even Intel's brand name and huge money can't buy everything.

So, these days Intel has only money and brand name and AMD has the best x86 CPUs ever.

Regarding Xiaomi, it has some cheap but very good smartphones and also some expensive ones but like AMD in sane prices for their hardware.

Intel and Nvidia are very well known of overpricing their products in insane levels, just because there are people with various psychological problems who actually buy them.

And because they have done everything they can - illegally - to eliminate competition and to become covered monopolies in order to sell at such high prices.

Now that Intel has competition after more than ten years, it has lost a good share in desktop and HEDT markets and next year is going to lose around 25% in server market.

But AMD is a small company and can't compete at the same time with nVidia and Intel.

Maybe the big failure of Turing and 7nm of Vega20 could make a difference.

We 'll see about that.

[Gser]

AMD has better CPUs now at all categories than Intel.
They are always faster on multithreaded apps and always cheaper.

Except for gaming and all other low threaded use cases. Even their Xenons have higher core clocks. I am not a fan boy of either just hoping AMD would step up their gaming performance of their CPU's and GPU's. Otherwise I'll just pay the nvidia and intel tax.

[huhn]

the gaming performance deficit comes mostly from the slower infinity fabric that has higher latency than the ring bus from intel.

that's why high speed low latency ram give an AMD more speed boost than an intel CPU because that closes the gap.

the clock speed difference shouldn't be anything more than a worse process and this should be fixed with TSMC 7nm and not a design problem from AMD.

especially for high 120+ FPS gaming intel CPUs are still far better and the reason i still don't have a new CPU...

[Blue_MiSfit]

Mind the toxic behavior, guys. Rule 15 and Rule 4 are both being broken.

[NikosD]

Especially for high 120+ FPS gaming intel CPUs are still far better and the reason i still don't have a new CPU...
So, you are telling us that the reason of not buying Ryzen is because it can play games at 1080p at 140 fps instead of 155 fps of Intel.

Interesting...

[huhn]

yes because your example is the difference between tearing and no tearing using a 144 screen run by fast sync and a ghosting free experience using BFI/back light strobing.

and the difference can be 20 % or more FPS.

[NikosD]

I knew that you would use the 144 Hz excuse, that's why I gave you that example...

So many people nowadays are using ultra expensive displays and ultra expensive GPUs in order to play games at 1080p resolution to reach 144 or more fps.

Right ?

[huhn]

144 hz screen have no special price tag any more and no you don't need expensive GPU it is relative easy to tweak the GPU usage but not the CPU usage at high frame rate.

and yes i personally plan to get a 240 hz screen.

[ShogoXT]

https://www.nvidia.com/content/dam/en-zz/Solutions/design-visualization/technologies/turing-architecture/NVIDIA-Turing-Architecture-Whitepaper.pdf

Bitrate efficiency and quality has gone up! This is a big deal see page 29.

[huhn]

"typical CPU"

there tests shows results where pascal is supposed to be close to x264 using PSNR. but it in term of visual quality it is not even close. x264 CPU utilisation 13 % good job nvidia...

and the 25% bitrate savings on HEVC well maybe they added b frames...

[NikosD]

Ok.

After all the reviews been released to the public, my original assumption still stands alive and kicking.

Turing RTX cards is the worst release ever for nVidia and maybe their biggest mistake.

I don't personally believe that RTX pseudo-raytracing is going to last if AMD doesn't support it.

Regarding RTX cards:

2080 Ti vs 1080 Ti is ~10% at 1080p, 25% at 1440p and 30% at 4K

2080 vs 1080 is ~ 35%

2080 vs 1080 Ti is almost a tie ~ 5%

1080 Ti had an MSRP around 50$ more than 980 Ti but ~ 70% more performance

1080 had an MSRP around 50$ more than 980 but ~ 60% more performance.

The actual difference in price of 2080 Ti vs 1080 Ti is ~ 500$

What a disastrous combination!

Stay away from this fraud.

[videoh]

Turing RTX cards is the worst release ever for nVidia and maybe their biggest mistake. It's so bad that no-one wants these things:

https://www.nowinstock.net/computers/videocards/nvidia/rtx2080ti/

/sarcasm

Take a look at the 5 year stock price history of nVidia to get the true picture.

I don't personally believe... Get over yourself, nobody cares.

NikosD can't afford the world's premiere GPU, so he goes the sour grapes route. What a joke!

[huhn]

so where is the competitions answer to this "fraud"?

it's not like people are happy about this release there is no alternative so they can do what ever they want. and it is not nvidias "propaganda" that created this situation is the fact that vega is a bad card.

[NikosD]

It's so bad that no-one wants these things:

https://www.nowinstock.net/computers/videocards/nvidia/rtx2080ti/

/sarcasm

Get over yourself, nobody cares.

Poor NikosD can't afford the world's premiere GPU, so he goes the sour grapes route. What a joke!Hahaha.

You are so funny guy!

nVidia was preparing for years this release with a leather-jacket-man show, a huge blog regarding RTX, a huge PDF file of 87 pages for...pre-orders.

Because after the release of real reviews, we had a world's first, the stock of the company with a huge marketing campaign to go down.

What a fail!

From your beloved pure capitalistic company Morgan Stanley with love:
We are surprised that the 2080 is only slightly better than the 1080ti, which has been available for over a year and is slightly less expensive.
With higher clock speeds, higher core count, and 40% higher memory bandwidth, we had expected a bigger boost.
https://www.techpowerup.com/247780/nvidia-stock-falls-2-1-after-turing-gpu-reviews-fail-to-impress-morgan-stanley

Poor videoh...Poor Turing...Poor nVidia.

It probably made its biggest mistake in recent history.

RTX for nVidia is like 10nm for Intel or MS Vista for Microsoft.

Big failures.

Let's see how nVidia is going to cope with this my poor friend.

[videoh]

Look at the 5-year plot. Stocks always vary over the short term. Your rhetoric and tone really expose you. "nVidia is a fraud" "leather-jacket-man" blah blah blah. Keep it up. You don't need others to discredit you.

When I get my pair of 2080 Ti's installed I'll think of you. :p

[NikosD]

No, stocks going down after real reviews is not a good sign at all.

But I never said nVidia is s fraud.

I said Turing is a fraud with that price for that performance.

Regarding leather-jacket-man, I don't like a guy presenting such a technical product to have a moto "the more you buy, the more you save"

But it seems he convinced you, at least.

So now you are definitely exposed as a blind nVidia fanatic.

What a pity :(

[huhn]

why don't you provide this blind person an eye opening alternative?

[NikosD]

Good news!

The worst stock price drop in the last 10 years for nVidia :)

https://www.techspot.com/news/77479-nvidia-stock-takes-nosedive-after-disappointing-earnings-report.html

[Atak_Snajpera]

Good news!

The worst stock price drop in the last 10 years for nVidia :)

https://www.techspot.com/news/77479-nvidia-stock-takes-nosedive-after-disappointing-earnings-report.html

Not so good if you are shareholder ;)

[foxyshadis]

We've officially entered another GPU recession; prices and availability will be better, but new designs and technology will be kicked another generation or two down the road, and investment will mostly center around refinements of current tech.

That's not an entirely bad thing, Nvidia's been repositioning itself as a luxury brand for too long now, and it will be nice to have affordable TFlops again.

[huhn]

the real problem here is the process. TSMC 12 nm "is" TSMC 16 nm so to make a faster cards that has place for special cores you have to make even bigger dies which result in higher prices the vram is more expensive too. nothing what people want but what we got the fact that AMD has no present with there vega cards doesn't make it any better.

so the question now is 7nm still to expensive to make smaller but faster GPUs or is nvidia trying to milk the end user here until AMD gets competitive again?

i don't understand how nvidia is a luxury brand when you get more or even performance for the same price as the AMD alternatives (atleast in the past two years).

and why does AMD need 2 different products for high end and main stream. you can downscale vega smaller than polaris much smaller but it doesn't look like it is worth it or where are the mainstream vega cards all these years?
but we got vega 64 which is far more expensive in production than a 1080 the HBM2 alone cost 3x GDDR5 and the die is bigger too...

[NikosD]

More and more I think that Raja Koduri doomed AMD GPUs with his Vega design.

He released a chip at 14nm with maximum CU (Compute Units) of 4096 and he couldn't match even 2nd place card of nVidia - Pascal 1080.

Now, even using 7nm, Vega can't upscale to more than 4096 CUs!

Practically we have to wait for another architecture from AMD to hit high-end cards and that would be probably the large Navi 20 at the end of 2019 (!)

If Raja wasn't paid by nVidia or Intel, he should.

Maybe that's why he is working officially for Intel now at discrete cards section.

He managed to ruin a generation of AMD's GPU cards in order for Intel to catch up and nVidia to try crazy things like pseudo-raytracing without consequences.

Anyway, I'm not interested in >1080p gaming so I'm going to wait for my 7nm RyZen 3000 series along with the small Navi 12 also at 7nm in Q2 of 2019.

[huhn]

well someone has to say it.

the major problem with RTX was that user didn't get more performance for there money. so now that AMD plans a card that does exactly the same thing by matching the price and performance with a 2080 which pretty much justified the RTX cards pricing...

worst possible outcome.

[nevcairiel]

The Vega architecture just wasn't that great. 300W on 7nm, with memory bandwidth to the moon and back, and they can only match a 2080, which does all of that on 12nm at 225W and only half the memory bandwidth. And not even for cheaper (MSRP is the same as 2080 non-FE)
How bad Vega was/is really shows with this release. Imagine what Turning could do with 300W on 7nm and that massive memory bandwidth.

The only people that might truely be interested in Radeon VII are possibly workstation users that can make full use of its 16GB memory and fast FP64 calculations.

Its a shame though that Navi just doesn't seem ready yet, and even if it was, its likely they are not going to launch anything high-end with it anytime soon, because that would directly undermine the Radeon VII.
Its no secret that I prefer NVIDIA cards at this point, but thats just a product of how underwhelming AMDs recent generations have been. And even if one isn't necessarily interested in buying an AMD card, competition drives innovation and keeps prices in check. And there just is none.

At least they managed to step back onto the CPU playing field, even though they are still playing catch-up in IPC performance. Hopefully the next 5-6 months of finalizing Ryzen 3000 can at least close the gap to Intel, and maybe even make a bit of headroom.
H2/Q4 2019 and 2020 will be interesting for CPUs. For GPUs, we'll have to see.

[NikosD]

Radeon 7 is an all around player.

Yes, it can play games around 30% faster than Vega so it matches 1080 Ti/ 2080 cards, but it can also be used as a workstation card.

It's not only the extremely fast FP64 performance, the out of this world memory bandwidth or the massive 16GB VRAM

It is also the drivers for professional apps, which should be similar to Vega Frontier.

So, a game dev for example could design and play his game on the same card, just by using a switch in the drivers.

HBM2 is still expensive obviously.

Now, Ryzen 3000 series shown at CES completely destroyed the competition.

A medium Ryzen 5 with only 8 cores, matched the performance of the top Core i9 9900K (!) using 130W vs 180W for Intel and without top clocks yet, because it was an engineering sample.

You should expect Ryzen 7 with 12 cores and Ryzen 9 with 16 cores.

Regarding Navi, it will come in summer like Ryzen 3000 series, smashing the middle category performance in terms of price/performance ratio.

Definitely this year belongs to AMD.

[huhn]

a german news site reported that the FP64 well be crippled.
and the 128 rop are wrong is the same ROP as mi50 64.

https://www.computerbase.de/2019-01/vega-20-amd-radeon-vii-fp64/

this is yet again a professional card salvaged for the consumer market.

[nevcairiel]

AMD wants to sell the MI50 after all, the only real reason to get it is the FP64 performance then, it's otherwise pretty close to a Radeon 7.

[Atak_Snajpera]

Personally I'm waiting for 75W Navi with RX580 performance.

[NikosD]

The last few days I'm under the impression that 1160 Ti or 1660 Ti or whatever name nVidia eventually decides to give to the Turing card with Tensor and RT cores disabled, will be a good card.

It seems to be a lot faster than 1060 6GB, RX 580 & RX 590, with a performance closer to 1070 - a little slower actually.

The initial estimation of its price is ~250$ and if that is true, I think I'm going to buy it.

I'll wait for Navi in summer along with Ryzen 3000 series for my new PC rig.

[huhn]

i highly doubt that a card with disabled feature will cost this much less the production cost is still the same.

if nvidia is able to produce new chip with no tensor or RT cores i can see a "huge" price difference.

but still if you don't need a new GPU right now i would wait for 7nm card which should give better performance for the money.

not sure if tensor cores should be disabled at all they will or should have a use case in "all" games at one point.

[clsid]

The chips are designed in such a way that broken parts can be disabled, while keeping a functional chip. So imperfect chips get sold as a lower model.

[huhn]

yes that'S known but the tensor core are a very specific part.

not like one of these core cuda packs where the defect can be at any of these.

[NikosD]

It seems that we only have to wait till February 15 and spend around 280$ in order to get a 1660 Ti.

Let's wait for reviews to find out if the performance is closer to 1070 or 1060.

https://www.tomshardware.com/news/nvidia-geforce-gtx-1160-1660-ti-specs-palit,38505.html

[wanezhiling]

I am interested in Intel's dGPU(Gen 12) in 2020

[NikosD]

Tough times for nVidia...But why ?

RTX is such a successful implementation with such a good price [emoji1787]

https://www.anandtech.com/show/13965/nvidia-earnings-report-q4-2019-crypto-pain-but-full-year-gain

[videoh]

Not so tough:

"For the full fiscal year though, earnings were still very solid, with revenue up 21% to $11.7 billion, and an overall gross margin of 61.2%, up 1.3% from 2018. Operating income was $3.8 billion, up 19%, and net income was $4.1 billion, up 36%. Earnings-per-share for all of 2019 came in at $6.63, 38% higher than 2018."

And don't forget nVidia also has about a 2 to 1 market share advantage over AMD, whose Radeon 7 has been a disappointment.

I'm loving my RTX 2080 Ti! The Turing SDK is out now and life is swell.

[huhn]

doesn't change that the company lost more than 30 % of it's worth in the last couple of month.

reviews and customer are not thrilled by these RTX cards.

I'm loving my RTX 2080 Ti! The Turing SDK is out now and life is swell.
this wasn't available on release date?

[nevcairiel]

Losses or not, without competition its not going to really matter, and AMD has only had very weak showings.

[huhn]

it clearly doesn't matter for the consumer and it shouldn't matter in the long run.

but if you are at the stock market than this is important. if is justified or not doesn't matter for the stock market.

[videoh]

Yeah, every time the nVidia stock price changes, the performance of my RTX 2080 Ti follows right along.

nVidia haters are getting desperate!

[huhn]

did you even read what i said?

i take it now as an compliant to be known as an nvida and AMD hater.

[videoh]

Haters gonna hate.

Work on your English and grammar; you're embarrassing yourself.

[huhn]

thank you for your contribution to this topic.

[NikosD]

One of the most terrifying moments that I recently had, was without doubt the reading of the following article:

https://www.tomshardware.com/news/softbank-sells-nvidia-shares,38564.html

SoftBank, a large investment fund, sold all of its nVidia shares - around 3.6 billion $ ! - due to low sales and no bright future ahead.

I have to repeat myself unfortunately.

Tough times for nVidia...

[Blue_MiSfit]

Let's keep it on topic, folks :)

[NikosD]

1660 Ti is eventually out as a new TU116 which is a Turing architecture die but without RT and Tensor cores.

The performance is close to 1070, so it's like 40% faster than 1060 6GB in games, but with the same TDP of around 120W.

With an MSRP price of 280$ it's definitely the most interesting Turing card to buy nowadays.

[huhn]

the 1660 ti give me a lot of question marks.

first is DLSS which pretty much lowers picture quality to increases performance so something i see as useful for a not high end GPU but it's missing i totally see why RT core are missing they are pretty useless on such a card but tensor cores?

next is the GPU size the relative TU116 size compare to a TU106 show it's only a little bit smaller per cuda core. the next thing are the design diagrams: https://www.pcgamesn.com/wp-content/uploads/2019/02/Nvidia-RTX-and-GTX-Turing-SM.jpg
so TU116 has dedicated FP16 cores where we can't see them on TU106 so the tensor core are doing the FP16 math?

[nevcairiel]

It's quite likely that the tensor cores are being recycled to do the FP16 math, it's far more likely then actually making new hardware.

The reason full tensor cores are disabled is the same old story, feature segmentation. They reckon gaming doesn't really need it, and they don't want some workstation loads going for cheap cards.

DLSS is specifically designed to offset RT cost. It's not designed to allow 4K gaming on low-end hardware. These are 1080p GPUs, the quality loss would probably be too extreme on that resolution (since it's more apparent in lower DPI)

[huhn]

i highly doubt DLSS is designed to off set RT it clearly very important for RT with it's very medicore performance but there are games that support DLSS but no RT like final fantasy 15 and DLSS can be used an alternative form of AA.
so even on a 1080p screen it has it's uses it just depends how it is used.

if i'm not mistaken metro exodus "struggles" to say it friendly with DLSS at "low" resolution but it wasn't used for super sampling in that game AFAIK.
The reason full tensor cores are disabled is the same old story, feature segmentation. They reckon gaming doesn't really need it, and they don't want some workstation loads going for cheap cards.
i totally understand your argument.
but i have to argue giving more user access to this specialised feature should create more program that use it and so more demand for these cards.

It's quite likely that the tensor cores are being recycled to do the FP16 math, it's far more likely then actually making new hardware.

i didn't knew they could do that i through they where extremely specialized in matrix operations.
i mean they can do some operation/outputs in FP32 so clearly not optimised for pure FP16

and the next thing is FP16 processing affecting tensor core performance unlike int32/FP32? if they are the reason turing can do great FP16 it has to be like this.
as a comparison polaris can do FP16 at the same speed as FP32 while pascal can do 1/64 of the FP32 so this is huge opportunity to write/modify programs that benefit from this.

(i take on paper numbers not real world test here)

[NikosD]

According to anandtech, based on nVidia sources the design of TU116 is unique regarding FP16 support, introducing for the first time dedicated FP16 cores.
Of course, as we just discussed, the Turing Minor does away with the tensor cores in order to allow for a learner GPU. So what happens to FP16 operations?
As it turns out, NVIDIA has introduced dedicated FP16 cores!

These FP16 cores are brand new to Turing Minor, and have not appeared in any past NVIDIA GPU architecture.
Their purpose is functionally the same as running FP16 operations through the tensor cores on Turing Major: to allow NVIDIA to dual-issue FP16 operations alongside FP32 or INT32 operations within each SM partition.
And because they are just FP16 cores, they are quite small. NVIDIA isn’t giving specifics, but going by throughput alone they should be a fraction of the size of the tensor cores they replace.

[huhn]

there is the next thing that puzzles me.
turing is supposed to be able to do integer and float at the same time but can it do FP16, FP32 and int32 at the same time? they are different cores aren't they?

[nevcairiel]

The normal Turing SM core can do int and FP at the same time now. On big-turning, FP16 can be done by the tensor cores, so it can also run in parallel - and on little-turing, the dedicated FP16 cores, if they are recycled tensor cores or not.
In Volta, the SM cores did FP16 still. This is also a Turing change.

I don't think they can tripple-issue commands to the same SM unit though, so it'll probably not actually do all three at the same time, but it may be flexible to do any two at the same time.

[huhn]

so none RTX and even some pascal cards can do ray tracing too?

i'm getting confused here nvidia you are sending me mixed messages here.

[NikosD]

so none RTX and even some pascal cards can do ray tracing too?



i'm getting confused here nvidia you are sending me mixed messages here.No confusion.

Microsoft's DXR API can be implemented in SW too besides specific ASICs, meaning that you can have ray-tracing using CUDA cores or Polaris/VEGA cores, even CPU cores.

The real issue here is speed.
Depending on the game, the speed is from 1.5 to 3 times slower between 1080 Ti and 2080.

nVidia allows Pascal (>1060), Volta (Titan V) and GTX Turing (1660/1660 Ti) to run RTX enabled games in SW (CUDA cores) using less rays/s to make RTX more popular in order to invest more on it.

[huhn]

no sorry i'm still confused by nvidia.

we have RTX card that get a new name to feature nvidia RTX.

then we get new turing card like the 1660 ti that can'T do RTX and can't do DLSS to help the FPS . these cards are supposed to have new FP16 cores which may or may to be tensor cores with disabled features.

and now we get an software update where and they can do nvidia RTX at a lower level but can't do DLSS. getting a lower quality/FPS for a lower tier GPU is to be excepted but what'S the pont of the name.

and i don't see how any of this change the real problem that nvidia can't fix the problem that the new RTX titles are not as popular as hoped by being "bad" games or by game studio marketing department shooting there self in the leg with an epic double barrel shotgun.

it was always know that the dx12 API can be used without RT cores the confusion comes from that the none RTX cards are supposed to use >RTX< ray tracing which was market as a special feature for the 2000 RTX series.

[nevcairiel]

RTX cards come with Ray-Tracing hardware, GTX cards come without it. I really don't see how this is that confusing.

And for Ray-Tracing popularity, it just takes time to develop. Just this month it was announced that both UE4 and Unity are gaining Ray-Tracing capabilities, which should open the market for a lot more potential games supporting it in the future.
Combine that with a much larger hardware-base (ie. Pascal card owners), and it might be far more common in the future to have at least basic effects that maybe even Pascal owners can use, and once you have that, the jump to advanced effects is only a small one.