Tom's Hardware Graphics 2012 Benchmarking Methodology Explained

Toggle sidebar Toggle sidebar
Stuka87

Stuka87

Diamond Member
Dec 10, 2010
6,240
2,559
136
I really do not think you can compare reviews from TH and AnandTech.
 
G

greenhawk

Platinum Member
Feb 23, 2011
2,007
1
71
I find it depends on what the "review" is about. too many of them are just cop-paste of marketing promotion flyers.

The rest are as interesting as any comparison can be. I still feel they are a poor shadow of what they use to be, espically when they started on the old sysdoc (?) server.
 
N

nyker96

Diamond Member
Apr 19, 2005
5,630
2
81
I personally like hard OCP reviews on g-cards. they use their own recorded benchmarks, not the official stuff that the driver may have optimized for and also they test per resolution and image quality which is more meaningful. It's not how much memory or Mhz you got, it's what image quality you can use in a particular game for a particular card that matters most. I think they are so honest they pissed off NV few years back, so they had to go out and buy their own cards from retail to do reviews. But that's a compliment to them not a complaint.

bet. Tom v. Anand, I'd go w/ Anand, just feels they are more fair in benching. but end of day hard OCP is the best for me.
 
B

blackened23

Diamond Member
Jul 26, 2011
8,548
2
0
utahraptor said:
http://www.tomshardware.com/reviews/graphics-card-benchmarks-charts-review,3154.html

I see a lot of comments in these forums and some others that Tom's Hardware has "canned" benchmarks or just biased practices in general. I always trust them the same as I trust Anandtech.

Do you think they are fair? Has this article changed your mind?
Click to expand...

He used override Edge Detect AA on the AMD cards, lol, Really? Yes, the 680 is faster and nobody here will dispute that (I don't think), but that reviewer is a freakin' moron.
 
MrK6

MrK6

Diamond Member
Aug 9, 2004
4,458
4
81
THG has always been a joke. No wonder their AMD results were so much worse than the competition in recent reviews.
 
Hallowed_Dragon

Hallowed_Dragon

Member
Feb 18, 2010
67
0
0
BallaTheFeared said:
They were probably just trying to equalize the IQ inequalities.
Click to expand...

Sure, sure... The quality of THG and their forums has become trash. The nVidia fanatics are roaring like crazy people there. I used to be a regular there, but I left when I saw the direction to which they were heading.
I'm glad I see some other people from there though (Hi notty)
 
S

Silverforce11

Lifer
Feb 19, 2009
10,457
10
76
BallaTheFeared said:
They were probably just trying to equalize the IQ inequalities.
Click to expand...

Since you seem to be in the know, can you show me an analysis of IQ at default settings in recent cards (any difference)? I just want to see your point of view on why you believe AMD offers unequal IQ.
 
B

boxleitnerb

Platinum Member
Nov 1, 2011
2,605
6
81
blackened23 said:
He used override Edge Detect AA on the AMD cards, lol, Really? Yes, the 680 is faster and nobody here will dispute that (I don't think), but that reviewer is a freakin' moron.
Click to expand...

Where exactly does it say that? Please point it out.

Edit:
I had a recent conversation with Igor and he told me that THG benches with current drivers for ALL cards and with currend game versions. This is not self-evident (although it should be)! Also I think it is commendable to explain the benchmark methodology - again, not everyone does that.
 
Last edited:
B

blackened23

Diamond Member
Jul 26, 2011
8,548
2
0
boxleitnerb said:
Where exactly does it say that? Please point it out.

Edit:
I had a recent conversation with Igor and he told me that THG benches with current drivers for ALL cards and with currend game versions. This is not self-evident (although it should be)! Also I think it is commendable to explain the benchmark methodology - again, not everyone does that.
Click to expand...

It says so on their graphs, and he confirmed it in an email to me. He was using override edge detect AA in some of the tests. Even with an apples to apples comparison, the best comparable AA mode nvidia has to it is CSAA and he doesn't use it on the nvidia cards to my knowledge. Even THEN, AA implementations vary substantially between vendors so you can't really do a comparison like that. It really should be left to application preference across the board.

Disclosing test methods is great and all, sure. I do think most reviewers include a "test setup" preface to include some of that information, though.
 
Last edited:
B

blackened23

Diamond Member
Jul 26, 2011
8,548
2
0
He only used it in some of the tests. It seems some of the graphs were modified, but he did use it in crysis 2.

I mean, how does that even work? Neither AMD (EQAA, AdAA, MLAA) or nvidia (CSAA, FXAA, TrAA) have the same AA modes generally speaking so how does one do a comparison between them.
 
Last edited:
B

boxleitnerb

Platinum Member
Nov 1, 2011
2,605
6
81
Okay let me get this straight. You (are you xoleras btw?) claim that he used override egde AA in some tests. There is no proof of this other "it seems" and the claim that "he confirmed it in an email to me".

Forgive me if I'm quite sceptical. Also I think you confuse AMD's edge AA with the one from Crysis 2 (Crysis 1 had it as well):



So basically you have absolutely nothing to show and call this reviewer a "freakin' moron". How am I doing so far?
 
Last edited:
A

amenx

Diamond Member
Dec 17, 2004
4,531
2,868
136
It seems THG is spoken of in the same tone here as AT is by AMDZone when it comes to CPU reviews. :p
 
C

Chiropteran

Diamond Member
Nov 14, 2003
9,811
110
106
The bitcoin mining results are so far from reality that it really puts into question all the rest of their benchmark results.
 
BrightCandle

BrightCandle

Diamond Member
Mar 15, 2007
4,762
0
76
There is a lot to get right on Video card reviews, and when it comes to AA its got a lot more complex in recent years. Toms isn't exactly the place to go for video card reviews, and hasn't been for years. But Anandtech isn't exactly top notch on this one either as it doesn't test surround/eyefinity. HardOCP test something I care about with best playable settings but it does so subjectively. Techreport do right with testing smoothness but don't test a variety of games and eyefinity or indeed crossfire/SLI on the recent cards.

No one has it right yet, and seeing as how everyone can see what the others are doing it amazes me they aren't all taking the best of each other and improving.
 
Olikan

Olikan

Platinum Member
Sep 23, 2011
2,023
275
126
BallaTheFeared said:
They were probably just trying to equalize the IQ inequalities.
Click to expand...

eb6.jpg
 
You must log in or register to reply here.

TRENDING THREADS

COMPANY
RESOURCES
FOLLOW
Top Bottom