6600xt vs 3060 , 1 month in!

[Phixsator]

So in order to disprove my comment you go ahead and post a benchmark where things are not set equally? Good job slick, you just earned the "Score in your own goal" award.

 

[Rroff]

So in order to disprove my comment you go ahead and post a benchmark where things are not set equally? Good job slick, you just earned the "Score in your own goal" award.

It is difficult to find newer results which compare both cards with a range of options (without DLSS skewing the picture) especially RT though I don't expect the story to have changed in that respect much since launch really.

 

[gpuerrilla]

I wouldn't expect a 3070 to run it well at 1440p at maxed settings. That's why I purchased a 3080.

What a silly answer. You argue about the 3060 then when pressed into the corner you dodge it. I think @Phixsator can throw back at you "I feel my original point has been proven here."

i want to make a clarification...3070 can't even reach 60 fps most of the time at native 1080p lmao at 1440p native with normal rt settings it wont punch anything above 40 fps unless you look at sky or be in nomad area...
its hilarious tbh. "good" rt performance yet its not. even a 3080 wont push a 60 fps at 1440p native with rt in cyberpunk.

Yep.

 

[gpuerrilla]

So in order to disprove my comment you go ahead and post a benchmark where things are not set equally? Good job slick, you just earned the "Score in your own goal" award.

Some guys go to extreme lengths just to prove another member wrong. Through gritted teeth just accept and move on.

The game doesn't run that great with RT features off either. Not sure what is up with the performance of the game for some people though

As he has said, its a poor title. You will need games that actually use both brands implementations that come out to get some better comparisons. Until then its all narrative.

 

[Nexus18]

Simple way to solve this, if amd cards are so good at raytracing now then getting posting in here:

https://forums.overclockers.co.uk/t...eature-test-port-royal-bench-thread.18838895/



Don't know why people can't accept the fact that RDNA 2 is just simply not as good as ampere for "ray tracing"...

Plenty of material out there showing the full picture and not heavily altered tests and extremely limited tests at that. Might as well have just reduced them to console settings and done a comparison with the ps5 and xbox...

Also, comparing to the tessellation fiasco with crysis 2 and witcher 3.... Not even in the same league, the heavy tessellation had literally no benefit to IQ where as ray tracing effects do.

That's the problem.. it isn't, RDNA2 ray traces at RTX2000 levels which was discounted by HUB along with RT on the 3000 series (remember the Nvidia debacle ?). Now AMD have the hardware they are limiting the cores of the 3000 series and suggesting that RDNA2 is equivalent.. terribly misleading.

When RDNA3 and rtx 4000 series release and RT gets taken to the next level the people who bought an RDNA2 card based on the above premise are not going to be happy.

And don't think that graphics are now limited to what a console can produce, that's a joke.

Good points there tbf.

People give a lot of **** to turing owners who upgraded to ampere cards for the increased ray tracing performance, wonder if we'll see the same come the next lot of amds cards

 

[Grim5]

Simple way to solve this, if amd cards are so good at raytracing now then getting posting in here:

https://forums.overclockers.co.uk/t...eature-test-port-royal-bench-thread.18838895/



Don't know why people can't accept the fact that RDNA 2 is just simply not as good as ampere for "ray tracing"...

Plenty of material out there showing the full picture and not heavily altered tests and extremely limited tests at that. Might as well have just reduced them to console settings and done a comparison with the ps5 and xbox...

Also, comparing to the tessellation fiasco with crysis 2 and witcher 3.... Not even in the same league, the heavy tessellation had literally no benefit to IQ where as ray tracing effects do.



Good points there tbf.

People give a lot of **** to turing owners who upgraded to ampere cards for the increased ray tracing performance, wonder if we'll see the same come the next lot of amds cards

There is some 6900xt cards there, their RT performance I slightly better than a 2080ti

 

[Joxeon]

That's the problem.. it isn't, RDNA2 ray traces at RTX2000 levels which was discounted by HUB along with RT on the 3000 series (remember the Nvidia debacle ?). Now AMD have the hardware they are limiting the cores of the 3000 series and suggesting that RDNA2 is equivalent.. terribly misleading.

When RDNA3 and rtx 4000 series release and RT gets taken to the next level the people who bought an RDNA2 card based on the above premise are not going to be happy.

And don't think that graphics are now limited to what a console can produce, that's a joke.

HUB like tossing out the occasional click bait video, the previous one was the OMG a 65w TDP CPU is only running at 65w on cheap B560 boards unless you change the power limit yourself.

It could have been presented as informative video to show how you can override the default spec and boost CPU performance but was instead painted like Intel have failed as these 65w don't come automatically unlocked on the really cheap boards.

 

[Phixsator]

Simple way to solve this, if amd cards are so good at raytracing now then getting posting in here:

https://forums.overclockers.co.uk/t...eature-test-port-royal-bench-thread.18838895/



Don't know why people can't accept the fact that RDNA 2 is just simply not as good as ampere for "ray tracing"...

Plenty of material out there showing the full picture and not heavily altered tests and extremely limited tests at that. Might as well have just reduced them to console settings and done a comparison with the ps5 and xbox...

Also, comparing to the tessellation fiasco with crysis 2 and witcher 3.... Not even in the same league, the heavy tessellation had literally no benefit to IQ where as ray tracing effects do.



Good points there tbf.

People give a lot of **** to turing owners who upgraded to ampere cards for the increased ray tracing performance, wonder if we'll see the same come the next lot of amds cards

Your completely missing the mark here. Nobody has said that AMD has perfect raytracing performance that match the raw horsepower of Ampere. They are on gen one of their implementation. Nvidia is on their second. However, recent titles that doesn't overuse RT is very playable on AMD hardware. So yes it has improved. Then again, the first RT titles was made through the nvidia developer "take money, use blackbox software with zero compatibility, gimp game" program aka GimpWorks. Didn't the 2080ti have horrible 1080p RT performance in battlefield to the point where DICE had to tone down the RT effects to get a workable result?

Heavy RT is the same as with tessellation, very little visual gain for a very steep performance price, ampere or not doesn't matter. Look at the 3080ti running Control 4k native High with only medium RT, still cant hit 60 and that game is so heavily sponsored by nvidia it's almost embarrassing it doesn't do better.

Might as well have just reduced them to console settings and done a comparison with the ps5 and xbox

You wouldn't want that as the 3060 would just take an even bigger beating . Ampere isn't known for being very scalable with performance at lower than 4k resolutions. But I'm sure you and the gang are going to make an argument for running these 1080p cards at 4k just to eek out a win, playability be damned .

 

[Nexus18]

Your completely missing the mark here. Nobody has said that AMD has perfect raytracing performance that match the raw horsepower of Ampere. They are on gen one of their implementation. Nvidia is on their second. However, recent titles that doesn't overuse RT is very playable on AMD hardware. So yes it has improved. Then again, the first RT titles was made through the nvidia developer "take money, use blackbox software with zero compatibility, gimp game" program aka GimpWorks. Didn't the 2080ti have horrible 1080p RT performance in battlefield to the point where DICE had to tone down the RT effects to get a workable result?

Heavy RT is the same as with tessellation, very little visual gain for a very steep performance price, ampere or not doesn't matter. Look at the 3080ti running Control 4k native High with only medium RT, still cant hit 60 and that game is so heavily sponsored by nvidia it's almost embarrassing it doesn't do better.

You wouldn't want that as the 3060 would just take an even bigger beating . Ampere isn't known for being very scalable with performance at lower than 4k resolutions. But I'm sure you and the gang are going to make an argument for running these 1080p cards at 4k just to eek out a win, playability be damned .

Key words there.... "that don't overuse", define overuse?

Just because amd/consoles can't handle it in the same way as nvidia can doesn't mean it is overused.....

Of course, both will still need FSR/dlss if you want to get much better/acceptable fps for a "pc experience" but this doesn't change the fact that nvidia handles more complex ray tracing far better than amd currently does, again, a good read:

https://www.pcgameshardware.de/Rayt...cials/ART-Mark-Raytracing-Benchmarks-1371125/

With every new ray tracing application, the promise of RT cores 2.0 and tensor cores 3.0 is put to the test again. The ART-Mark shows what Control, Minecraft RTX & Co. could only sketch out blurred, namely where the journey could go in the future: The more complex the ray tracing in space, the better Ampere performs. The extent of the difference depends on the implementation, with the ART mark being pushed to the limit with the help of Unreal Engine 4. The exaggerated use of rays at high settings to represent reflections in reflections, which cause a multitude of bounces - reflected light - shows the graphic potential of the technology. If you compare Turing-TU102 as 1st-Gen-RTX with Ampere-GA102, Ampere is up to a factor of 2.7 faster. If you only use the (halfway) smoothly displayed settings, in this case "RT High", it is still a factor of 2.2 and thus significantly more than in all current raytracing games. That is certainly due to the scene

If you compare Nvidia's first RT generation with AMD's first RT generation, the result is a tie. In fact, the RDNA-2 GPUs, better known as the Radeon RX 6000, can handle more complex ray tracing calculations better than Turing. The RT High preset of the ART-Mark runs just as well on a Radeon RX 6900 XT as it does on a Geforce RTX 2080 Ti. Compared to an RTX 3060 Ti and above, no Radeon herb has grown with raw ray tracing. AMD will tackle this backlog with RDNA 3, that's for sure.

No doubt amds first attempt is good if you want to compare to nvidias first gen of ray tracing but how long ago was turing released? And we are living in the present here, not the past.....

And nope, 3060/6600xt is not a "PC 4k card", want a true 4k pc experience, 3080/6800xt or higher is the way to go. I rather get a ps5/xbox than a 3060/6600xt card.

 

[K1LaW8tt5]

I wouldn't expect a 3070 to run it well at 1440p at maxed settings. That's why I purchased a 3080. I almost went 3090 but sanity told me that I wouldn't notice a difference at said resolution. Looking at the heat/power draw issues I'd say I dodged a bullet.

a 3060/3060ti/3070 would have been a better choice for you to pair with your "old ivy bridge system". Bit of a waste buying a 3080.

Imagine talking about console/AMD performance limitations when you do things like that.

 

[Wrinkly]

What a silly answer. You argue about the 3060 then when pressed into the corner you dodge it. I think @Phixsator can throw back at you "I feel my original point has been proven here."

Read previous posts.

 

[gpuerrilla]

Read previous posts.

No u

 

[Wrinkly]

a 3060/3060ti/3070 would have been a better choice for you to pair with your "old ivy bridge system". Bit of a waste buying a 3080.

Imagine talking about console/AMD performance limitations when you do things like that.

Few things -

• Do you think I wish to remain with a 3770k now that software is starting to tax it?
• Do you think a 3060/3060Ti/3070 will out perform a 3080 with a 3770k.
• Do you think a 3060/3060Ti/3070 will out perform a 3080 with an up to date CPU?
• Do you think the best place to promote consoles is an enthusiast PC forum?

 

[Wrinkly]

No u

I made my point regarding a misleading video.
Poster is then mislead by video.

What is your problem now?

 

[Alan McFall]

So, if a popular hardware reviewer tested the 11700k vs the 5950x, limited it to 8 cores and then claimed that the multicore performance was fairly even without testing all core performance you wouldn't see that as misleading in any way ?

 

[Joxeon]

And nope, 3060/6600xt is not a "PC 4k card", want a true 4k pc experience, 3080/6800xt or higher is the way to go. I rather get a ps5/xbox than a 3060/6600xt card.

The cards are ok it's the price which is the issue, one should cost £250 and the other £280.

 

[nvidiamd]

Simple way to solve this, if amd cards are so good at raytracing now then getting posting in here:

https://forums.overclockers.co.uk/t...eature-test-port-royal-bench-thread.18838895/



Don't know why people can't accept the fact that RDNA 2 is just simply not as good as ampere for "ray tracing"...

That test is more close to the Minecraft type of RT and not CP or ME which are still hybrid RT games. A more appropiate test to judge the hybrid RT performance is the Port Royale benchmark and yes Ampere is like 25% ahead in there.
But this has nothing to do with the low end 3060 or 6600xt, these cards are not able to run any form or RT at high settings/resolutions. HU were right to reduce some settings to the point where the game becomes playable, there is no reason to show that a card gets 10 fps but it is still better than the other that gets 5 fps.

 

[K1LaW8tt5]

Few things -

• Do you think I wish to remain with a 3770k now that software is starting to tax it?
• Do you think a 3060/3060Ti/3070 will out perform a 3080 with a 3770k.
• Do you think a 3060/3060Ti/3070 will out perform a 3080 with an up to date CPU?
• Do you think the best place to promote consoles is an enthusiast PC forum?

The first 3 are questions you should have asked yourself before buying a 3080.

• Maybe you don't, but if you upgrade in 12 months that's almost 2 years of wasted GPU power and console like performance limitations.
  
• No, but at 1440p (and DLSS) the gap will be much smaller in open world true NeXt GeN games.
  
• No, but you don't have an up to date CPU. If you did you would get the full benefit of the 3080.
  
• No.

glad I could help

 

[Wrinkly]

That test is more close to the Minecraft type of RT and not CP or ME which are still hybrid RT games. A more appropiate test to judge the hybrid RT performance is the Port Royale benchmark and yes Ampere is like 25% ahead in there.
But this has nothing to do with the low end 3060 or 6600xt, these cards are not able to run any form or RT at high settings/resolutions. HU were right to reduce some settings to the point where the game becomes playable, there is no reason to show that a card gets 10 fps but it is still better than the other that gets 5 fps.

 

[pieman109]

I replaced a reference sapphire 6800 (non xt) with a 6600xt pulse, and to be quite honest I don't notice any difference. This is more than likely down to me playing only older games (COH: opposing fronts, Homeworld remastered, and most of the COD series as well) as newer games don't interest me. All games played at 1440p, most settings on high and the only time it struggled was when mayhem was going on in game and the fps only dropped to around the high 40s.
Newer games would be a different story I reckon but I'll just replace the 6600xt as when required (if suitable alternatives are available, so thats never!)