AMD Radeon VII Retested With Latest Drivers

[xkm1948]

I suggest you relax a bit xkm1948 as you have been writing constant provocative posts for hours in this thread. And I think that might end bad for you. A simple advice. I understand your pain or anger for your FuryX not being what you wanted to be but this attitute in an AMD thread towards anyone liking their products is doing harm for sure. And you use to do that a lot for months now in any AMD GPU related thread. Don't take it as a personal attack, just pointing out that it is a dangerous path for a TPU forum member the one you follow lately.

Suggesting people to buy and test themselves is provactive? Since when I can no longer participae in discussions? They say what they wanted to say, i suggest they use their own purchase to backup their claim and that is wrong? Are you trying to censor postings now?

 

[EarthDog]

I suggest you relax a bit xkm1948 as you have been writing constant provocative posts for hours in this thread. And I think that might end bad for you. A simple advice. I understand your pain or anger for your FuryX not being what you wanted to be but this attitute in an AMD thread towards anyone liking their products is doing harm for sure. And you use to do that a lot for months now in any AMD GPU related thread. Don't take it as a personal attack, just pointing out that it is a dangerous path for a TPU forum member the one you follow lately.

The ignore list is brand loyalty agnostic...

Really, there are some pretty scary opinions on both sides of the fence here that need reigned in or just straight kicked out. If my ignore list his 15 (and lord knows we are close!), I'm just walking from this fookn forum.

 

[Vayra86]

Suggesting people to buy and test themselves is provactive? Since when I can no longer participae in discussions? They say what they wanted to say, i suggest they use their own purchase to backup their claim and that is wrong? Are you trying to censor postings now?

Well I have to agree it does all sound a bit baity and it doesn't help the overall quality of topics. That is at least what I associate this with. Its not like you are genuinely interested in the answers. Consider it feedback...

 

[notb]

Please explain the 55 cents, I'm intrigued

Well, the calculation was very long - I don't want to bore anyone with the details...
But a few variables cancelled and I got this in the end:
exp(2*ln("current year" - 2009) + "which planet from Sun you're on") = exp(2*ln(10)+3)
that's the way you calculate how much a GPU should cost, right?

And it may actually explain @Super XP 's results, because, judging by his comments, he's clearly from a different time or planet...

 

[xkm1948]

Well I have to agree it does all sound a bit baity and it doesn't help the overall quality of topics. That is at least what I associate this with. Its not like you are genuinely interested in the answers. Consider it feedback...

Dude I am ALL for scientific methods. I am a molecular genetist. You make a hypothesis and you back it up with experimental setup and results. Anyone can go around all day and fart claims out lf thin air. But to make solid claims one has to get their hands dirty and get to do the experiments: FineWine as one of many examples.

 

[Vayra86]

Dude I am ALL for scientific methods. I am a molecular genetist. You make a hypothesis and you back it up with experimental setup and results. Anyone can go around all day and fart claims out lf thin air. But to make solid claims one has to get their hands dirty and get to do the experiments: FineWine as one of many examples.

The point is, those examples exist and we already know what's up. Stop beating the dead horse...

 

[EarthDog]

Well why not try a new occupation at AMD and contribute in making their gpus better instead talking how shait their hardware is.

Sorry, you will need to get in line behind the dude from Adored who nearly lost his gag reflex trying to get a job at AMD.

 

[notb]

Well why not try a new occupation at AMD and contribute in making their gpus better instead talking how shait their hardware is.

You do understand that molecular biology and semiconductor engineering are 2 different things, right?
But man... wouldn't it be nice if AMD finally employed (or at least cooperated with) people that represent the "pro" client segment.
Nvidia is *so active* in promoting and supporting GPGPU and it clearly didn't hurt their products...

When looking at recent AMD products and how they try to be attractive to content creators, it does remind me of how open-source software for photographers looks.
Software people are like: "we gave you all these fantastic features! Why are you still using that dreaded, expensive Photoshop?!"
And AMD is like: "we gave you all this fantastic cores and we crush Intel in benchmarks! Why are you still using that awful, expensive i7?!"

Sorry, you will need to get in line behind the dude from Adored who nearly lost his gag reflex trying to get a job at AMD.

Man... I'm actually thinking about applying for a job at Nvidia. I wonder if I'd still be able to take part in all these lovely flames if they choose me... :/

 

[turbogear]

@W1zzard
Thanks a lot for the effort to retest.
Me as Radeon VII owner was not expecting any large performance increase between 19.2.1 and 19.2.2. The time frame I believe is too short between the two.
It did though manage to fix few issues with Wattman and overclocking.

What I personally noticed is that Black Ops 4 is running better. With 19.2.1 on my PC the frame was changing very fast between 70 FPS and 120FPS by every movement in game. The GPU clock was having lot of dips. With new drive it is mostly varying above 100FPS staying close to 120FPS and GPU clock is relatively stable. I have BenQ Zowie XL2730 FreeSync 144Hz monitor.

 

[HD64G]

Dude I am ALL for scientific methods. I am a molecular genetist. You make a hypothesis and you back it up with experimental setup and results. Anyone can go around all day and fart claims out lf thin air. But to make solid claims one has to get their hands dirty and get to do the experiments: FineWine as one of many examples.

Since you are fond of numbers that prove things scientifically, here is an example that proves that AMD GPUs aren't shown in their best form at their launch and this helps us customers to get an equal or better product in better price than will deserve in a few months only. And for a customer that keeps his hardware at least for 3 years, this is an opportunity.

 

[Nkd]

I'm simply stating facts. I would wish that AMD GPU's were able to compete.. All their "high-end" solutions have been terrible recently. Leaving this segment for Nvidia.

Not sure what you mean. Does this card not compete at all? AMD needed CPUs to survive so they focused on that. Or you would have no AMD and Radeon graphics. Now CPU side is going good they will eventually get to the GPU side. Given their need to survive as company I am not sure how this is so bad. They somehow managed to stay relevant which in its own is good enough while starting to give intel some serious competition on CPU side. AMD did what they had to and thought long term. If they invested more in GPU side they might not be in business right now because CPU side is where they are going to survive as a company. Its not a matter of if, its just when. I think in next few years will have their Zen of GPUs too.

 

[trog100]

well i am glad i forked out for a 2080ti.. at least i dont have to worry about the rest..

i follow all this stuff because it interests me.. but its all being spoiled by to much negativity.. too much entitlement infringement.. too much silly red green bickering..
huge lengthy threads about f-ck all...

trog

 

[xkm1948]

Since you are fond of numbers that prove things scientifically, here is an example that proves that AMD GPUs aren't shown in their best form at their launch and this helps us customers to get an equal or better product in better price than will deserve in a few months only. And for a customer that keeps his hardware at least for 3 years, this is an opportunity.

This is exactly my point. For people who don’t believe in W1zzard’s review, go get a Radeon 7. Test it along the life span of the GPU and report back how it ages.

 

[Countryside]

This is exactly my point. For people who don’t believe in W1zzard’s review, go get a Radeon 7. Test it along the life span of the GPU and report back how it ages.

W1zzard’s reviews are of top quality and i dont doubt his ability but he is one, most people dont have the expertice nor the equipment to test hardware thus they realy on reviews and you dont make an educated decision based on one review.

 

[siluro818]

I huv the money for a top of the line peformer. U mad cause i can afford something good? lol

Also i am not your bro, kiddo

Oh nobody's arguing that you have the money, clearly lacking in every other area though

 

[notb]

The point is, those examples exist and we already know what's up. Stop beating the dead horse...

While I'm not a huge fan of the argument quality we've seen here, I do share @xkm1948 's enthusiasm for a more scientific approach (as shown earlier).
I mean: we've been hearing about FineWine for years and it doesn't look like this could ever stop. There are, as you said "examples" or "traces" that something could be going on, but that's a bit like a proof of existence of ether or gods - we've seen some weird stuff going on, so let's assume there's a reason and let's give it a catchy name.

We had a nice discussion about quality of journalism in another topic. Seriously, wouldn't you like to see a proper experiment for this theory?
It's not that hard either. You take a few cards from different generations, 20-30 driver editions released over 5 years, a few AAA games and you measure all data points.
Instead, we have hundreds of thousands of people on thousands of forums talking about a phenomenon that no one has ever confirmed in a controlled environment. And it's been going for years. And it will keep going for as long as AMD will be making GPUs.

Why hasn't any of the large tech sites done this? Are they afraid of the result and angry mob from either side?
Why hasn't AMD done this?

 

[siluro818]

Wait, wait, wait: "I am a molecular genetist."

*nondescript snorting sounds*

Oh god that was just... Try to put together some molecules for that blood pressure, mate.

 

[xkm1948]

While I'm not a huge fan of the argument quality we've seen here, I do share @xkm1948 's enthusiasm for a more scientific approach (as shown earlier).
I mean: we've been hearing about FineWine for years and it doesn't look like this could ever stop. There are, as you said "examples" or "traces" that something could be going on, but that's a bit like a proof of existence of ether or gods - we've seen some weird stuff going on, so let's assume there's a reason and let's give it a catchy name.

We had a nice discussion about quality of journalism in another topic. Seriously, wouldn't you like to see a proper experiment for this theory?
It's not that hard either. You take a few cards from different generations, 20-30 driver editions released over 5 years, a few AAA games and you measure all data points.
Instead, we have hundreds of thousands of people on thousands of forums talking about a phenomenon that no one has ever confirmed in a controlled environment. And it's been going for years. And it will keep going for as long as AMD will be making GPUs.

Why hasn't any of the large tech sites done this? Are they afraid of the result and angry mob from either side?
Why hasn't AMD done this?

HardOCP did. They found Nvidia does not gimp their cards and AMD does not FineWine either

 

[HammerON]

Warning and reply bans issued. No more personal attacks. Play nice or move along.

 

[notb]

HardOCP did. They found Nvidia does not gimp their cards and AMD does not FineWine either

OK, I've just checked it. Not enough data points, wrong presentation. But otherwise the right direction. Thanks.

 

[bug]

OK, I've just checked it. Not enough data points, wrong presentation. But otherwise the right direction. Thanks.

Wrong attitude I'd say. You're looking for someone to disprove by the book something that has only anecdotally been proven.

 

[Mistral]

If I may, a suggestion for benchmarks going forward: have a small icon next to the game titles, indicating if they are nVidia or AMD sponsored.

 

[xkm1948]

Long term performance, people dont usually keep their cards long enough to see the impact. With the progression rate most modern GPUs have usful gaming performance life span for 3~4 years. By the time FineWine is ready the GPU is already obsolete.

 

[Countryside]

If I may, a suggestion for benchmarks going forward: have a small icon next to the game titles, indicating if they are nVidia or AMD sponsored.

That's a decent idea, or instead of the icon you can just add text next to the game title nVidia or AMD sponsored

 

[lexluthermiester]

The improvements offered by these updated drivers are on par with NVidia's same efforts for RTX and seem to show that with proper optimizations Radeon 7 is going to be a great card for a ton of usage scenario's.