Very interesting to see people call out all the bullshit this go-round, including W1zard's test suite amongst other things (like the actual real price during it's relevance).
It's not just the games, which I've long-explained are biased, but it should include things like mins in RT/upscaling, not just averages for RT (so people can see limitations/real use-cases).
This is very important.
I also know it's also a shit ton of work, but that's where we're at, because it just is. We need to see 1440p and 4k 'quality' and 'performance' upscaling w/ and w/o rt, and their minimums. Also raster.
You have to evaluate 7900xtx and cards oriented towards RT and/or greater than one level of up-scaling on their own merits (including quality of that up-scaling). You shouldn't simply conflate the two.
This is non-negotiable for relevance levels of important. If you want to split them into a different graph or set of graphs, that's fine, but that information needs to be there.
You can't just use average mins. You can't just separate RT averages. It is so incredibly misleading to people I don't even know what to say. I have to believe he knows this?
On top of that the title is just...wow. That's a new level,
even for this site's typical commentary. Was that comparison in the reviewer's guide? I'm pretty sure it truly was for the 4080s (also misleading).
Good job, every-men. Carry on. More people are starting to get it (and asking the right questions)! Hooray! This is why you should never rely on just one source, especially this one...not anymore.
Few issues down that people are starting to absorb, only a whole bunch left to go. Like I've been trying to explain, testing needs to be completely revamped...especially on this
very important site.
It's not just the feature-set, but also ram limitations shown through IQ differences that are not reflected in frame-rate; more visual, as we talked about
here.
I want W1zard to have the credibility he deserves due to his work/datasets (all the info; like clocks/voltages at different uses, fan info, etc), but his games tested/methods/averages info need a
serious revamp.
Not doing this really does damage his cred, imo; it does not lead to an accurate analysis for consumers. Given he's one of the few written/static sites with this much info, it's important it stays relevant/accurate.
Does he want that, though? I honestly don't know anymore, and haven't for a while.
Prove me wrong: that's what I'm saying. I'm not insulting the man; just prove me wrong by being better, then we all win.
Then I can shut up about all this stuff because people will actually see it on a grand scale rather than me having to explain it.
I could PM some suggestions for potential improvements, iyw? I have many beyond mentioned here that would make things more applicable to more people and their general uses; give a clearer picture.
Not trying to be aggravating, unthankful, or biased in either direction; I just truly want this to be a reliable source of pertinent information. TPU is becoming less X-bit Labs and more...something else.
Honestly, chaps in this forum, I'm really proud of you all for calling this out en masse. I really, truly, am.
I don't honestly know if this is an honest misrepresentation or W1zard has ulterior motives, but I can tell you how it's looked in the past and continues to look with this review.
Apparently, now, so can many other people.
If you want to 'ignore' people calling this out, that's fine. Revel in your own ignorance and accept the not-currently-shown limitations/advantages brought about by your choices, for that is your right.
At some point someone will breakthrough and more will understand the important divide that is happening (between raster and rt/up-scaling inc IQ), and how things are shifting in perhaps unexpected ways.
Maybe not today, but someday.
Some already do, and I'm happy for that.