People that think TPU is biased towards Nvidia tend to be biased towards AMD. It's like a faith system, where a result you do not like must be attributed to falsehood.
Given TPUs preponderance to 'grey market' key resellers, it shows that it's unlikely to be taking backhanders from Nvidia.
As an additional factor, it's clearly obvious more people dislike Nvidias business direction, myself included, whereas I'd argue it's hard to 'hate' on the underdog but that's human nature.
We live in an age of confirmation bias where fanboi's of any side of an issue only want to read what fits their preconceived notions. I do read other sites for some info not available or not always available here, but i have no interest or even understand any reason to use "benchmarks" as a measurement criteria in the context used. It's very easy for a driver author to tweak the benchmark to excel at what the benchmark includes at the expense of things it doesn't. As for business direction, welcome to capitalism. Boards are required by law to prioritize sgarehoilder interests / profits within the confines of the law. They get no points for being "nice guys". The public hammered nVidia for their partner program to their own detriment. Under the program, "parners" would get various nVidia incentives which many of us read as "marketing speak" for "become a partner and when we detect the partner card, we'll loosen up the restrictions in Boost 3:
AMD does play the role of the nice guy being the company that cares about the average Joe. Folks hammer nVidia for the cost of G-Sync because "Freesysnc is free and it's the same thing" which is about as true as AWD is the same as 4WD ... which as anyone who regularly drives in snow or has gone off road will just laugh. With today's upper tier cards, most games are well out of the range < 70 - 80 fps range where Any-Sync shines ... having the option to use MBR technology which requires a hardware module that Freesync does not have is "not the same thing".
I understand that when buying a car, if the model with AC is $200 more, choosing the one without AC cause 'you only need it 30 days a year". I don't understand the argument to justify buying the one without AC when both models are the same price which is what we saw with PhysX. We are now getting the same argument with RTX, you don't think RTX is worth the performance penalty ? Turn it off.
My biggest frustration with AMD GFX is if you continue to promise and don't deliver what everyone was expecting, you're hurting yourself. Gotta get off this train:
1. Prelaunch hype is all [Insert new AMD technology here] is going to change everything. (Fill in HBM, Mantle whatever)
2. Benchmarks "leaked" <cough, cough> showing it beats comparably priced nVidia card in [this one thing]
3. Card is released at price similar to nVidia card and almost as fast using hand picked games ... "but ,,, ooh ooh ... with overclocking it can match nVidia !"
4. Testing of cards post release shows, 1) yeah it is almost as fast in those games, further behind in most others. 2) It needs more power, emits more heat and is louder than competition's cards, 3) It needs a 50 - 100 watt bigger power supply and an extra case fan.
5. Further testing shows not only what we already knew ... nVidia cards can be overclocked too, but also that nVidia cards can be overclocked more, even widening the gap.
6. Pricing drops and now the comparison isn't so bad ... would have been a lot less "ill will" if they'd just started there in the 1st place.
On that last note .... it does seem that nVidia is doing a bit of imitation of AMD ... Since 2xx series, AMD has been very aggressively overclocking the card in the box which is why they jumped to 95C operating temps at that time. That left very little room for manually overclocking with increases in fps averaging 4 - 6%. The green team was almost always in double digits improvements with some breaking 30%. With the Radeon VII (131.0 fps), we saw TPU get 8.2% after manual OC ... it was supposed to compete with the 2080 but TPU managed just 9% with the FE 2080 (169.0) making the 2080 FE 29% faster in the manually overclocked test. The 2070 FE (138.7) is what it should be more properly compared with the 2070 FE OC'd 8.1% giving it a 4.5 % speed advantage.
It would seem that nVidia spoiled expectations with the large boost over 9xx performance such that they chose to release the 2xxx series with far less manual OC ability putting them in the boxes with a pretty aggressive OC as compared to past generations. But as to the business practices, this is Macy's versus Gimbels, Ford versus Chevy, etc. I don't see either being more consumer friendly than the other, they are just playing the cards that were dealt. You do what you can with what you got.
if AMD released the Radeon card at 2070 pricing, the reaction would i think be positive.
Now, how does this affect the OPs issue ... there's two parts of TPUs test procedure that favor AMD:
1. I don't think there's a large % of users here who aren't using MSI AB to OC their cards. I wish there was a way to click on or check a box that showed two different graphs. te default is what we have now ... the optional would be click this and see them compared with their respective OCs. Say for example in the OC page we see that the card being tested OCs 8.2%. Click on the compare OC'd option and that 8.2% is added to the scores in that graph. Now is this valid ? More so than I originally thought I'd imagine. Yes all card scores vary depending on what each game entails, but outside any frame limits, I would think that if a particular game gains 10%, since it's dealing with the same loads before and after the manual OC, gains should not vary a lot. To test each game individually would be to much to ask of the reviewer. It would be nice to see an article where all 23 games in the test suite were tested with the manual OC so we could see just how wide the variance is as a matter of reference.
What I think would be a huge feature is, taking advantage of what I imagine exists ... database with all cards performance for each game would be a "compare" function where you could click to compare the card being reviewed and ask it to add any card which had been tested with that came and same "test system". So for Radeon VII, I'd see a list of say 20 cards and one might pick 5 I wanna see compared. So a table with 7 colums would be on screen .. Game name card in article and the 5 other cards I picked. The deafult view could be at 1440p w/ radio buttons to switch to 1080p and 2160p and default being "out of box" speeds and another pair to pick out of box ormanually overclocked with the latter just being "the math" applied from above. Would be a great tool assuming the assumption holds reasonably close.
2. The cost per dollar thing. To my eyes, choosing of xx80 over a xx70 on a cost per performance barometer can not be made using the cost of the card alone. The entire system is involved in delivering that user experience. So if the 2070 is $550 (and everything else is $1300), then when compared against the 2080 ($790) build the cost ration is not 790 / 550 (44%), it's 2090 / 1850 (+12.9%). To be fair... if it means a 650 PSU instead of a 550 and an extra case fan, then that needs to be allottedfor also.
I see your point. I guess what leaves me off-put is the constant comparison to Nvidia. Which is really my problem. I have no interest in Nvidia or their products...for personal reasons. Thus...I don't really care how other products stack up compared to theirs. I don't care how theirs perform. I don't care how much theirs cost. I don't care if for the same price theirs is better. I don't care if for a lesser price theirs is better. I JUST DO NOT CARE ABOUT NVIDIA. But I realize other people do. And that for a review something usually needs compared to something else to make certain points about it. I just wish it wasn't always a comparison to Nvidia somethings. Is that sensible or reasonable? Probably not. None-the-less...I don't like it.
Long story short...it's me. Not you. And I know that. But I still gotta be me.
That's certainly a valid position ... from your point of view no one could argue differently, but the question then becomes "of what purpose is the review ?" I you want the best performance you can get from a vendor you can live with, then just buy AMDs top card, no need to read a review. No reason to care if it doesn't stack up well. Personally, I tend to be a "hardware whore", the vendor that gves me the best performance, within my budget" gets my buisiness. There are other factors ... from 2002 up till I guess the Z87 / 7xx series era, we were almost exclusively an Asus shop; if a user didn't have a preference, we'd go w/ Asus MoBo and GFX. They almost always had a performance edge, their BIOS was complete, familiar and easy than competition and support was great. Then all that changed. We continued to use a lot of monitors, mice and routers. Now that's changed. Our last RMA took 3 months. Now the only response we get from Asus TS, when we get any are non-responsive tot he question asked. It's like the tech , instead of going thru his 3-ring binder to find the question, and providing the answer is just throwing the binder up in the air and giving us the response on the page that the binder was open to when it landed on the floor. Asus is on our "no buy" list for everything as of this week due to lack of responses with a weird mouse issue and 3 router issues. This was the proverbial last straw.
In order to be prepared for questions when someone wants a new build, I try and keep appraised of strengths and weaknesses. I use the TPU summary charts as a source of an overall performance gradient but when auser has interest only in certain games, will confine the analysis to just those. From that starting point, I look at manual OC results and apply those to the TPU charts. I look at power consumption, noise heat and consider anything which might over ride the performance comparison. Our build philosophy is to never have a repeat customer so we teach folks how to do their own component selection, build the box and maintain it. That means starting with a base build for each budget, then listing oither options for each selection. So a build that say has a Scyther $45 Fuma cooler ... options with equivalent performance that might be considered are these 3 air coolers and thse 3 CLCs, go home, read about their pluses and minus, come back and let's discuss.
I have had favorites... Muskin was my go to source fro RAM for about 20 years .... always top end performance, unrivaled ability to handle crazy OC voltages and TS was just one level below coming to your house. But with DDR$, I have not been impressed with performance or availability and we've parted ways. Humans tend to look at the world strictly from their own Ponit of View (PoV). Henmce the well discussed philosophical question "If a tree falls in the forest and nobody hears it, did it make a sound ?" The scientific mind says "of couse it did,, the laws of science are not suspended just because there's no on there to hear it ... he defines sound as something made". The philosopher / humanist says no, ... "since no one heard it, it is of no significance, he defines sound as something heard". So when we have had 1, 2 or even 3 bad experiences with a particular brand, we think of ouselves as significant enough to matter, we aren't ... statistically speaking our experiences are too few to form any reliable conclusion. Look how many folks swear that they will never own another hard drive of a certain brand ...( I had 3 failures !)... and yet, when we look at RMA rates, the differences are insignificant. Last 12 month period for which data available:
- HGST = 0.975%
- Seagate = 0.825%
- Toshiba = 0.930%
- Western = 1.15%
That being said, every vendor comes out with crap models ... if your Model has over 2.0%, look elsewhere. These were the worst 5
- 10,00% Seagate Desktop HDD 6 To
- 6,78% Seagate Enterprise NAS HDD 6 To
- 5,08% WD Black 3 To
- 4,70% Toshiba DT01ACA300 3 To
- 2,95% WD Red 4 To SATA 6Gb/s
In the end the value of a review is the information in it. Not the conclusions, not the recommendations, just the numbers and the cost. Let's say there's someone out there who plays only Witcher 3 and also won't use nVidia. The Radeon VII hits 95.4 fps @ 1440p ... his monitor isn't Freesync or G-Sync but it is equipped wit MBR capability (BenQ Zowie 27 inch 144Hz Gaming Monitor, (Model XL2735). At 95 fps with MBR a radeon VII, he'll have a better experience than an equivalent Freesync Monitor or G-Sync user who doesn't or can't use ULMB ... so all is right with the world and nothing else matters.. For the fellow or gal who is open to more options, they have the option to get 109.2 for the same price. Will they have a noticeably better experience .... I sincerely doubt it. But if they read the review and know a little math ....
a) They will know that the 2080 will be about 57% as loud.
b) They can use a 80 - 100 watt smaller PSU
c) At average USA electric cost they will save about $18 a year, if they live by me, that's about $39
d) They will need 1 less case fan
Perhaps not of interest to everyone, but these are some of the reasons why i read TPU reviews.