Jump to content

Rumor? nVidia RTX 2080 & 2080 Ti Reviewers Guide Leak? Lots of 4K

WMGroomAK
Just now, Lathlaer said:

These are 2 different games.

woops , quoted the wrong graph , look at these numbers 

image.png.941a67167954910371db0f7764ef1889.png

RyzenAir : AMD R5 3600 | AsRock AB350M Pro4 | 32gb Aegis DDR4 3000 | GTX 1070 FE | Fractal Design Node 804
RyzenITX : Ryzen 7 1700 | GA-AB350N-Gaming WIFI | 16gb DDR4 2666 | GTX 1060 | Cougar QBX 

 

PSU Tier list

 

Link to comment
Share on other sites

Link to post
Share on other sites

1 minute ago, Space Reptile said:

woops , quoted the wrong graph , look at these numbers 

Does Battlefield 1 have a built-in benchmark? 

 

If not then there is nothing surprising with having different numbers for the same game. Testing place, testing duration, it all matters. What's important is the relative performance in the same test.

 

For instance, Shadow of the Tomb Raider has a built-in benchmark and I can confirm that what's on NVIDIA's graph here more or less matches 1080ti independent result. 

CPU: i7 6950X  |  Motherboard: Asus Rampage V ed. 10  |  RAM: 32 GB Corsair Dominator Platinum Special Edition 3200 MHz (CL14)  |  GPUs: 2x Asus GTX 1080ti SLI 

Storage: Samsung 960 EVO 1 TB M.2 NVME  |  PSU: In Win SIV 1065W 

Cooling: Custom LC 2 x 360mm EK Radiators | EK D5 Pump | EK 250 Reservoir | EK RVE10 Monoblock | EK GPU Blocks & Backplates | Alphacool Fittings & Connectors | Alphacool Glass Tubing

Case: In Win Tou 2.0  |  Display: Alienware AW3418DW  |  Sound: Woo Audio WA8 Eclipse + Focal Utopia Headphones

Link to comment
Share on other sites

Link to post
Share on other sites

2 hours ago, Taf the Ghost said:

GeForce-RTX-2080-Ti-RTX-2080-R6S-1000x447.png

 

This one stuck out to me the most.

 

https://www.pcgamer.com/rainbow-six-siege-benchmarks/

 

EYNDHVXiKN4D5DMeDdFQBa.thumb.png.acda03cebc40abbe05269bb62ff9cb8b.png

 

This is February 2018 testing, so it's recent. We've got the 1080 at 44 FPS and the 1080 Ti at 59 FPS. How the hell did Nvidia's marketing get those numbers?

At 1440p Ultra + 4x T-AA the Rainbow Six Siege consumes 7,5GB of VRAM and gets me well over 100FPS (with 1080ti).

There is no way that this was run on 4k with a 4x T-AA and the amount of VRAM consumed at those settings would hit the VRAM limit on the RTX 2080.

Link to comment
Share on other sites

Link to post
Share on other sites

13 minutes ago, WereCat said:

At 1440p Ultra + 4x T-AA the Rainbow Six Siege consumes 7,5GB of VRAM and gets me well over 100FPS (with 1080ti).

There is no way that this was run on 4k with a 4x T-AA and the amount of VRAM consumed at those settings would hit the VRAM limit on the RTX 2080.

3qTc5C2jUcKA2MHzTkc5Ba.thumb.png.f214dc4268ea25ad52cc3b77fe659cf9.png

 

PCGamer backs up your numbers, just as a note.

 

It would appear Nvidia took the texture quality way down, and found a way to make it more of a Memory Bandwidth test?

Link to comment
Share on other sites

Link to post
Share on other sites

@Space Reptile

 

Here you have Tom's Hardware: 

 

aHR0cDovL21lZGlhLmJlc3RvZm1pY3JvLmNvbS9X

 

 

And here is Rainbow Six from TechPowerUp

 

r6siege_3840_2160.png

 

 

For every review with one sets of FPS there is at least one other with something different. 

CPU: i7 6950X  |  Motherboard: Asus Rampage V ed. 10  |  RAM: 32 GB Corsair Dominator Platinum Special Edition 3200 MHz (CL14)  |  GPUs: 2x Asus GTX 1080ti SLI 

Storage: Samsung 960 EVO 1 TB M.2 NVME  |  PSU: In Win SIV 1065W 

Cooling: Custom LC 2 x 360mm EK Radiators | EK D5 Pump | EK 250 Reservoir | EK RVE10 Monoblock | EK GPU Blocks & Backplates | Alphacool Fittings & Connectors | Alphacool Glass Tubing

Case: In Win Tou 2.0  |  Display: Alienware AW3418DW  |  Sound: Woo Audio WA8 Eclipse + Focal Utopia Headphones

Link to comment
Share on other sites

Link to post
Share on other sites

3 minutes ago, Taf the Ghost said:

3qTc5C2jUcKA2MHzTkc5Ba.thumb.png.f214dc4268ea25ad52cc3b77fe659cf9.png

 

PCGamer backs up your numbers, just as a note.

 

It would appear Nvidia took the texture quality way down, and found a way to make it more of a Memory Bandwidth test?

Also, Rainbow Six Siege by default sets resolution scaling to something like 70%-75% even if it detects your HW and sets the other settings and resolution correctly. It is maybe possible that the 4k 100+FPS results are with a resolution set to 4k but the res scaling to be bellow 100%.

Link to comment
Share on other sites

Link to post
Share on other sites

3 minutes ago, Lathlaer said:

@Space Reptile

 

Here you have Tom's Hardware: 

 

aHR0cDovL21lZGlhLmJlc3RvZm1pY3JvLmNvbS9X

 

 

And here is Rainbow Six from TechPowerUp

 

r6siege_3840_2160.png

 

 

For every review with one sets of FPS there is at least one other with something different. 

There's drivers and settings issues, but no one has gotten any R6S to run anywhere close to 80 FPS on a 1080 or or 100 FPS on a 1080 Ti.

Link to comment
Share on other sites

Link to post
Share on other sites

16 minutes ago, WereCat said:

At 1440p Ultra + 4x T-AA the Rainbow Six Siege consumes 7,5GB of VRAM and gets me well over 100FPS (with 1080ti).

There is no way that this was run on 4k with a 4x T-AA and the amount of VRAM consumed at those settings would hit the VRAM limit on the RTX 2080.

It's very possible it isn't doing any AA at all.  And with better memory bandwidth it is possible it doesn't need that much ram.

Grammar and spelling is not indicative of intelligence/knowledge.  Not having the same opinion does not always mean lack of understanding.  

Link to comment
Share on other sites

Link to post
Share on other sites

24 minutes ago, Taf the Ghost said:

There's drivers and settings issues, but no one has gotten any R6S to run anywhere close to 80 FPS on a 1080 or or 100 FPS on a 1080 Ti.

 

23 minutes ago, mr moose said:

It's very possible it isn't doing any AA at all.  And with better memory bandwidth it is possible it doesn't need that much ram.

OK so I fired up Rainbow Six just for a quick test on 1440p.

 

1440p

100% res scaling

x4 T-AA

HBAO+

Everything else maxed out

20180915134138_1.thumb.jpg.58066e8688dceb6e72a92b645497a37b.jpg

 

I got around 38-45FPS and the VRAM usage was at 6,7GB of VRAM (in the single-player, in-door map).

 

1440p

100% res scaling

x1 T-AA

No AO

Shadows on High instead of Very High

20180915134627_1.thumb.jpg.064c8f446bb032505ecb426079fa2bb7.jpg

 

I got 140-200FPS and the VRAM usage was at 4,7GB in the same map.

 

image.png.d353ebd4f5fdd1a24e1615783972e0e7.png

 

So my mistake claiming that x4 T-AA can run above 100FPS (definitely not) But everything Ultra + HBAO+ and x1 T-AA definitely can reach 100FPS (at 1440p, not 4k!)

 

EDIT:

Just for clarification 1x T-AA still uses at least 6GB of VRAM at those settings.

 

Link to comment
Share on other sites

Link to post
Share on other sites

Comparing different test results does not add up, because FPS is not just GPU alone.

We had that discussion on some German pages that used a Threadripper system and got WAY lower GPU FPS as a result.

 

CPU matters is the takeaway. You can't compare results, without making sure the whole test system is identical. Including the cooling, open / closed, room temperature, etc.

 

The Same GPU with a strong Intel chip will be significantly higher than with an equally priced AMD chip. That is pretty common knowledge I thought. 

How do you guys think a CPU Gaming Benchmark is done? There are huge differences.

Link to comment
Share on other sites

Link to post
Share on other sites

7 minutes ago, Tech Enthusiast said:

There are huge differences.

Not in 4K. I don't know any game where the CPU choice in 4K would result in drastic changes. Some, yeah, but usually no more than single digit FPS. 

 

It's not like the outlets tested 1080ti with a Pentium CPU. It's usually, depending on the time of test, a 6700k, 7700k or 8700k. None should provide meaningful difference in 4K.

CPU: i7 6950X  |  Motherboard: Asus Rampage V ed. 10  |  RAM: 32 GB Corsair Dominator Platinum Special Edition 3200 MHz (CL14)  |  GPUs: 2x Asus GTX 1080ti SLI 

Storage: Samsung 960 EVO 1 TB M.2 NVME  |  PSU: In Win SIV 1065W 

Cooling: Custom LC 2 x 360mm EK Radiators | EK D5 Pump | EK 250 Reservoir | EK RVE10 Monoblock | EK GPU Blocks & Backplates | Alphacool Fittings & Connectors | Alphacool Glass Tubing

Case: In Win Tou 2.0  |  Display: Alienware AW3418DW  |  Sound: Woo Audio WA8 Eclipse + Focal Utopia Headphones

Link to comment
Share on other sites

Link to post
Share on other sites

12 minutes ago, Tech Enthusiast said:

 

The Same GPU with a strong Intel chip will be significantly higher than with an equally priced AMD chip. That is pretty common knowledge I thought. 

How do you guys think a CPU Gaming Benchmark is done? There are huge differences.

I wouldn't say equally priced, equal pricing more often leads to the AMD CPU being the better product particularly for demanding CPU games but it's just in general any Intel CPU will get higher outright max FPS figures in low GPU demand tests. There's just lots of existing games that just run better on Intel no matter the CPU, you have to go really garbage Intel CPU before it flips over to an AMD CPU performing better in those kinds of tests.

Link to comment
Share on other sites

Link to post
Share on other sites

26 minutes ago, leadeater said:

I wouldn't say equally priced, equal pricing more often leads to the AMD CPU being the better product particularly for demanding CPU games but it's just in general any Intel CPU will get higher outright max FPS figures in low GPU demand tests. There's just lots of existing games that just run better on Intel no matter the CPU, you have to go really garbage Intel CPU before it flips over to an AMD CPU performing better in those kinds of tests.

Well, possible yes. I don't follow CPUs as closely as I follow another tech. Just kinda think CPUs are boring *cough*.

 

The point I wanted to make is: Don't compare results if you don't know they were done on the same system.

Link to comment
Share on other sites

Link to post
Share on other sites

I don't get why people are still comparing the 1080ti with the 2080ti. With this release NVIDIA x-ed the initial Titan card for enthusiasts. Therefore the Titan Xp successor is the 2080ti. The 2080 goes against the 1080ti (as always – why would you buy a new more expensive card if the top of the line card from last gen is barely slower and cheaper?).

So the equivalene looks like this:

Titan Xp <–> RTX 2080 Ti ($1200 vs. $1200-ish, both from NVIDIA and MSRP)

GTX 1080 Ti <-> RTX 2080 ($699 vs. $799 both NVIDIA FE, $699 MSRP for 3rd party cards, will take a while to reach it)

GTX 1080 <-> RTX 2070 ($549 vs. $599- for NVIDIA FE or $499-ish for 3rd party cards)

GTX 1070 <-> GTX 2060 (…)

GTX 1060 <-> GTX 2050 (…)

GTX 1050 <-> don't bother (…)

 

The first leaked scores of the 2080 Ti suggest similar performance with non-release drivers in not optimized scenarios as a Titan Xp without making use of the new technologies. The 2080 as of now costs about 14% more than a 1080 Ti FE. So it needs to improve performance for at least 14%. Anything on top makes it a better deal. Don't compare 3rd party cards that are now (meaning 1 1/2 years after launch!) cheaper than MSRP with pre order prices at launch. That's just utterly stupid. The RTX 2070 FE costs just ~9% more than a 1080 while offering new technologies. If it's at least 9% faster it's a better deal. Again: forget preorder prices for 3rd party cards. The original 1080 was released over 2 1/2 years ago, the updated one 1 1/2 years ago. Plenty o time for the market to settle a bit. 

 

To summarize it, the RTX cards need at least the following performance boosts to make sense:

RTX 2080 Ti: 0% compared to Titan Xp

RTX 2080: 14% compared to GTX 1080 Ti

RTX 2070: 9% compared to GTX 1080

 

It's fair enough that people might be a bit disappointed because of the lack of price to performance drops after previous launches but if the new cards still offer that kind of performance they are objectively a better deal. If they don't than NVIDIA fucked up. 

Use the quote function when answering! Mark people directly if you want an answer from them!

Link to comment
Share on other sites

Link to post
Share on other sites

22 minutes ago, bowrilla said:

It's fair enough that people might be a bit disappointed because of the lack of price to performance drops after previous launches but if the new cards still offer that kind of performance they are objectively a better deal. If they don't than NVIDIA fucked up. 

The notion that after waiting 2 years for new architecture people should be happy with 0% FPS boost because of raytracing and DLSS is, quite frankly, ridiculus. 

 

More importantly, this is not how you do business. NVIDIA is in the business of selling GPU's and they would not be able to sell a single Turing GPU to a whole bunch of segment (Pascal owners) if that would be their goal. 

 

Somehow they could retain similar prices between 780 and 980 and 1080 and yet the performance boost was not 0% there. 

 

I get it, there is ray tracing and there is DLSS. But you have to put some kind of value on those technologies and that value is not absolute. There is no scenario in which an owner of a Titan Xp will consider 2080ti as a "successor" to their GPU if it offers 0% FPS boost, raytracing notwithstanding.

CPU: i7 6950X  |  Motherboard: Asus Rampage V ed. 10  |  RAM: 32 GB Corsair Dominator Platinum Special Edition 3200 MHz (CL14)  |  GPUs: 2x Asus GTX 1080ti SLI 

Storage: Samsung 960 EVO 1 TB M.2 NVME  |  PSU: In Win SIV 1065W 

Cooling: Custom LC 2 x 360mm EK Radiators | EK D5 Pump | EK 250 Reservoir | EK RVE10 Monoblock | EK GPU Blocks & Backplates | Alphacool Fittings & Connectors | Alphacool Glass Tubing

Case: In Win Tou 2.0  |  Display: Alienware AW3418DW  |  Sound: Woo Audio WA8 Eclipse + Focal Utopia Headphones

Link to comment
Share on other sites

Link to post
Share on other sites

18 minutes ago, Lathlaer said:

There is no scenario in which an owner of a Titan Xp will consider 2080ti as a "successor" to their GPU if it offers 0% FPS boost, raytracing notwithstanding.

I highly doubt there will not be an FPS increase going from Titan Xp to the 2080ti.

i9-9900k @ 5.1GHz || EVGA 3080 ti FTW3 EK Cooled || EVGA z390 Dark || G.Skill TridentZ 32gb 4000MHz C16

 970 Pro 1tb || 860 Evo 2tb || BeQuiet Dark Base Pro 900 || EVGA P2 1200w || AOC Agon AG352UCG

Cooled by: Heatkiller || Hardware Labs || Bitspower || Noctua || EKWB

Link to comment
Share on other sites

Link to post
Share on other sites

1 minute ago, TahoeDust said:

I highly doubt there will not be an FPS increase going from Titan Xp to the 2080ti.

Me too. I was just responding to the previous poster and his assumptions. 

CPU: i7 6950X  |  Motherboard: Asus Rampage V ed. 10  |  RAM: 32 GB Corsair Dominator Platinum Special Edition 3200 MHz (CL14)  |  GPUs: 2x Asus GTX 1080ti SLI 

Storage: Samsung 960 EVO 1 TB M.2 NVME  |  PSU: In Win SIV 1065W 

Cooling: Custom LC 2 x 360mm EK Radiators | EK D5 Pump | EK 250 Reservoir | EK RVE10 Monoblock | EK GPU Blocks & Backplates | Alphacool Fittings & Connectors | Alphacool Glass Tubing

Case: In Win Tou 2.0  |  Display: Alienware AW3418DW  |  Sound: Woo Audio WA8 Eclipse + Focal Utopia Headphones

Link to comment
Share on other sites

Link to post
Share on other sites

17 hours ago, asus killer said:

i don't see performance increase that justifies the price difference. Only raytracing does, if there is games using it and you care for said games

Hehe no one does - probably why this now exists xD 

https://www.slashgear.com/nvidia-scanner-rtx-gpu-overclock-in-one-click-14545709/

 

I once did the unthinkable, back many headphones ago...

I split an audio split, again

Link to comment
Share on other sites

Link to post
Share on other sites

4 minutes ago, Lathlaer said:

Me too. I was just responding to the previous poster and his assumptions. 

Well, congratulations, in that case you didn't understand what I was saying. Since the Titan Xp and the 2080 Ti have the same MSRP the 2080 Ti only needs to be on par to be a better deal since it comes with new technologies. May it be disappointing? Sure. It still is a better deal based on the facts and it doesn't matter if people feel that way or not. The same performance with the option for raytracing, tensor cores that make DLSS available. If you already have a Titan Xp then sure, it doesn't make much sense. If you don't and are in for a new enthusiast grade top of the line GPU then it surely is an alternative. And people who are in for the bragging rights will buy it anyways. And you can expect the other cards to deliver otherwise NVIDIA has some serious sale issues. Maybe they're trying to keep the prices up as long as they have 10 series stocks, who knows. The preorder prices are irrelevant though and you can't compare them with the current street prices.

Use the quote function when answering! Mark people directly if you want an answer from them!

Link to comment
Share on other sites

Link to post
Share on other sites

23 minutes ago, bowrilla said:

Well, congratulations, in that case you didn't understand what I was saying.

No, I think I understood you just fine. You seem to have mistaken disagreeing with you with not understanding you.

 

You called 2080ti a successor to Titan Xp based on MSRP with which I can even agree. What I will not get behind is the incredibly low performance increase bar you have set for 2080ti. 0% based on MSRP and raytracing is not acceptable for me as an Pascal owner. 

 

There is a difference between being a successor and being an alternative. That new tech combined with hypothetical 0% improvement makes it an alternative to Pascal, not a successor. 

 

Ano what people like me want from this premiere is progress, not an equally priced alternative to our cards. This seems to be where our opinions differ.

CPU: i7 6950X  |  Motherboard: Asus Rampage V ed. 10  |  RAM: 32 GB Corsair Dominator Platinum Special Edition 3200 MHz (CL14)  |  GPUs: 2x Asus GTX 1080ti SLI 

Storage: Samsung 960 EVO 1 TB M.2 NVME  |  PSU: In Win SIV 1065W 

Cooling: Custom LC 2 x 360mm EK Radiators | EK D5 Pump | EK 250 Reservoir | EK RVE10 Monoblock | EK GPU Blocks & Backplates | Alphacool Fittings & Connectors | Alphacool Glass Tubing

Case: In Win Tou 2.0  |  Display: Alienware AW3418DW  |  Sound: Woo Audio WA8 Eclipse + Focal Utopia Headphones

Link to comment
Share on other sites

Link to post
Share on other sites

26 minutes ago, Lathlaer said:

You called 2080ti a successor to Titan Xp based on MSRP with which I can even agree

No, not just considering MSRP. NVIDIA usually released a new Titan model along with a new generation. The later Ti version beats both the Titan and the regular non-Ti version. That usually gets people annoyed as well. This time there is no enthusiast Titan, it's just the RTX lineup and the Titan V that clearly isn't marketed for gaming at all. A probable explanation is that NVIDIA just got rid of the enthusiast gaming Titan cards.

 

26 minutes ago, Lathlaer said:

0% based on MSRP and raytracing is not acceptable for me as an Pascal owner.

Fair enough. For you as a Pascal Titan Xp owner this is not interesting. So just skip it and be happy that you don't have to spend 1200 bucks? I do understand that you're disappointed if it was only 0% increase but we don't know that yet – even though I don't expect more than maybe 10% in current titles the don't utilize the new technology.

 

26 minutes ago, Lathlaer said:

There is a difference between being a successor and being an alternative. That new tech combined with hypothetical 0% improvement makes it an alternative to Pascal, not a successor

Since it comes after Pascal to market and is aiming for the same niche it is a successor. It would be an alternative if both cards would be on the market at the same time. I expect NVIDIA to sell the last stocks and stop production.

 

26 minutes ago, Lathlaer said:

Ano what people like me want from this premiere is progress, not an equally priced alternative to our cards

The RTX line up provided progress for the first time in years if not decades. (Partial) real time raytracing is a huge step forward. It might not be visible right now at this moment but this is the next step and a wet dream of developers for years.

P.S.: Just have a look at CPUs. There's rarely a point at which it makes sense to switch directly for the next generation. If you already had a 6700k there was basically no point upgrading to a 7700k. If you had a 1700X there's barely a point to upgrade to a 2700X. If you have a Titan Xp there's barely any point ugprading to the next enthusiast level card other than bragging rights.

Use the quote function when answering! Mark people directly if you want an answer from them!

Link to comment
Share on other sites

Link to post
Share on other sites

2 hours ago, bowrilla said:

Just have a look at CPUs. There's rarely a point at which it makes sense to switch directly for the next generation. If you already had a 6700k there was basically no point upgrading to a 7700k.

This is my point exactly. People were not exactly thrilled with the situation on the CPU market. It's not something to strive for.

 

Meanwhile, there was always a compelling reason to upgrade to the next GPU architecture. If you had a 780ti, upgrading to 980ti gave you tangible results. If you had a 980ti, upgrading to 1080ti was a significant boost as well. 

 

I am not saying that raytracing is garbage - it's not. It's a wonderful piece of technology that is badly needed in times where people tend to chase FPS and higher "K" in resolution only. BUT - given that 1) not every game will support it and 2) games that do support it will probably require heavy compromises on other settings - I am saying that in this early shape and form, it's not worth eshewing raw power altogether. I am saying that those cards - the first cards that feature this new hardware - need to be able to give than just raytracing in order for the Turing release to be called successful. 

 

And I'm not alone in this camp. Once people learned that raytracing might require heavy blows to the other aspects of experience (lower resolution, fewer FPS) - they jumped onto the rhetorics that since the new tech might not be applicable (especially with cards lesser than 2080ti), they want raw FPS as well. Only the problem is that while the raw fps gain may be visible when comparing two cards of the same nomenclature (2080ti vs. 1080ti), it's not so sexy when comparing two cards of the same MSRP (2080 vs. 1080ti). 

 

All things considered, we will see in couple of days.

 

I am reserving my final opinion and purchasing choices for when I get the full picture - unfortunately during latest WAN show Linus heavily implied that no official benchmark at this point will include raytracing performance simply on the account of it not being implemented in time.

CPU: i7 6950X  |  Motherboard: Asus Rampage V ed. 10  |  RAM: 32 GB Corsair Dominator Platinum Special Edition 3200 MHz (CL14)  |  GPUs: 2x Asus GTX 1080ti SLI 

Storage: Samsung 960 EVO 1 TB M.2 NVME  |  PSU: In Win SIV 1065W 

Cooling: Custom LC 2 x 360mm EK Radiators | EK D5 Pump | EK 250 Reservoir | EK RVE10 Monoblock | EK GPU Blocks & Backplates | Alphacool Fittings & Connectors | Alphacool Glass Tubing

Case: In Win Tou 2.0  |  Display: Alienware AW3418DW  |  Sound: Woo Audio WA8 Eclipse + Focal Utopia Headphones

Link to comment
Share on other sites

Link to post
Share on other sites

Let's note - not sure its' been said yet - Nvidia just delayed the RTX2080 ti to the 27th general public release. Pre-orders are now slated to arrive betwen the 21-27th, with no mention of release day delivery available. I have mine ordered with overnight, and it's showing delivery is scheduled for the 24th. 

SLI 1070 Setup for sale!!!

Link to comment
Share on other sites

Link to post
Share on other sites

10 hours ago, WereCat said:

 

OK so I fired up Rainbow Six just for a quick test on 1440p.

 

1440p

100% res scaling

x4 T-AA

HBAO+

Everything else maxed out

 

I got around 38-45FPS and the VRAM usage was at 6,7GB of VRAM (in the single-player, in-door map).

 

1440p

100% res scaling

x1 T-AA

No AO

Shadows on High instead of Very High

I got 140-200FPS and the VRAM usage was at 4,7GB in the same map.

 

 

 

So my mistake claiming that x4 T-AA can run above 100FPS (definitely not) But everything Ultra + HBAO+ and x1 T-AA definitely can reach 100FPS (at 1440p, not 4k!)

 

EDIT:

Just for clarification 1x T-AA still uses at least 6GB of VRAM at those settings.

 

So the difference between  (4xAA and HBAO+) and (1xAA and No AO) is basically a 1-2 G of ram and about a 100 -140 FPS.  If that is the case then it is conceivable that with the new 2080 not having to run any AA at all on it's cuda cores that it will perform like the 1080ti with no AA but it will be running DLSS on the tensor cores.  

 

Which just leaves the questions as:

1. will DLSS be good enough to replace current AA?

2. Is there even a need for it?

3. if so will people accept that the added tensor cores and associated performance effects (assuming they are positive) are legitimate performance upgrades or will they get stuck looking at the number of cuda cores and ram size and complain it's not worth the money?

 

 

Grammar and spelling is not indicative of intelligence/knowledge.  Not having the same opinion does not always mean lack of understanding.  

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

×