THG Busted Photochopping CPU pics?

[pHaestus]

http://arstechnica.infopop.net/OpenT...4&m=8400979235

The evidence seems pretty convincing to me (Photoshop newbie though)

[mo]

Im extremely disappointed. At best, it was a very biased article. At worst its a complete dud. Im more than a little peeved that the folx at THG. Its no excuse that Frank Völkel wrote it. No decent company will publish an editorial without approving it! This is really making me reconsider having my next system be a P4. No matter what, one needs to consider the kind of company his money is supporting:P

[pHaestus]

I wouldn't blame intel for the lameness of Tom.

[TonyM]

Just wonderin has Frank Völkel or THG said anything in response to these findings?

[Joe]

hehehe I didnt know people read that site anymore

[DarkEdge]

I read the site only when they do their big Mobo roundups. I see what they come up with. Take their top 3, then compare their top 3 and see where they stacked up in other editors reviews. System works pretty good for me.

As for creating the photoshop image...well thats lame!

[g.l.amour]

is there anyone in here that knows fotoshop well enough to say its true or not?

dunno who to believe, amdzone might not be the best to point the scam out.

[Brad]

1. Take what you think is copied.
2. Overlay it over the copy.
3. Perform a subtract. As both copies will be identical, the subtraction will result in perfect black - Usually a box.



hat monster is right, that image is definately photochopped

[iceheart]

Using the difference layer blending mode is a hell lot faster than performing a subtract.

[jamicon]

i just used copied it to a new layer, then moved it over the other 3, when lined up i kept toggling between visible and hidden for that layer, apart from the colour being slightly different they are exactly the same.

[gmat]

Ow, come on, it's so easy bashing THG. Please refrain from bullying them or i'll feel compelled to defend those ridiculous, pompous and lifeless idiots.
Oops i did it.
Nahh really what was in their minds when they decided to edit that photo ??
Let's state the facts
- They OC'ed a P4 to unholy speeds -> with proper cooling and power feeding, i don't see what's wrong there
- They decided to put up the results in a provoking article, just to get bashed by acneic and immature AMD fanboys -> again i don't see whats wrong there
- They decided to give some life to the article with a "gutter press headline" plot -> that's their very problem, what counts is the results. It's not a good point for a pretending-to-be-serious review site but well...
- To enlighten it some more they added a faked photo, seemingly to sustain the "stolen P4 sample" plot idea -> The problem is, it's aint obvious. Say, if they had painted the number with a big red brush, no one would have complained and eveeryone would have understood the initial idea.
Now the result is, all this looks like a fake. The plot is fake, the photo is a (non-first-sight-noticeable) fake, so why the whole article wouldnt be a fake ?
Say, if they edited the P4 photo in a way no one would notice at first sight, why didnt they fake bench results, POST photos and so on ?

Thats it, gutter press sensationalism killed the last milligram of credibility they still had.
Yes, i was still reading their GPU or Mobo round-ups, the photos and graphs were good.
The TH News section was lost since they started copy/pasting PR releases.
Now the reviews section is lost as well.
What's left ? A forum polluted polluted with kiddies and idiots.. not much.

It seems TH died a 9/11 as well... what a sad day.

[mo]

I ran across this when I read through THGC forum, which really bothered me:
http://www.faceintel.com/

[tim-may]

i usually cruised THG every few months or so, and i had read the 3ghz p4 review the day after it was posted. something did seem a little fishy about the POST screen, but i couldnt quite figure out what it was. seems it was right to suspect something...

[ymboc]

Just a heads-up....

they posted an excuse/explanation for the pictures...

http://www17.tomshardware.com/cpu/02...4_3600-17.html

*shrug*

-ymboc

[TonyM]

Kind of a shady response, but thats just my take on it.

[tim-may]

shady indeed.....

[WebMasta33]

You know what the funny thing is?

Aren't the codes on the CPUs traced, so intel could know, espically with those engineering samples, who has those CPUs?

I think if that's possible, they're kinda fuxored right about now.

[bigben2k]

My take on this is that THG did get a pre-engineering sample, but they got it from someone else that has an NDA (non-disclosure agreement) with Intel: it would explain why they hide the sample#.

The thing is, it's got the wrong core, as it should have a Prescott core, and that's why the markings don't make sense, and why everyone is questioning it.

I think Intel shipped "whoever" an engineering sample of a 3.3 MHz P4, which ended up in THG's hands, but then Intel decided that they wouldn't use the old core when they go to 3.3. Why they decided to do this would make up a good article.

And that makes the whole article pretty much pointless, because they're talking about something that will never make it to market. Also, if you look at the Athlon comparison benchmarks, they weigh pretty heavily on the Intel side, where even an Athlon 3000+ doesn't beat a P4 2.53. That seems awfully skewed to me.

It seems clear to me that THG has a connection with "someone", and they're using this connection to stay alive. THG articles (to me) don't have much technical merit, but they do have some value, for example, in introducing new products.

THG also seems to depend on its registered members to produce some kind of marketing survey, and that's got to be a money maker.

I'm sure that if THG could get their hands on AMD engineering samples, they might start publishing articles that have fair comparisons. THG has been accused many times before of favoring Intel, but if they try to be fair, then they'd be independant, and probably wouldn't be making as much money.

THG has an integrity issue and flawed technical reviews, but THG won't go away, not anytime soon, because it's probably a cash cow for Tom, wether you visit the site or not, which you probably do, if you favor Intel, and have a narrow minded view on the issues, or simply for the other non-CPU articles that they have, which are more interesting (but still flawed).

Does anyone know how many sites covered this controversial THG article? That must have generated quite a lot of traffic to THG... Money in the bank, for Tom.

[pHaestus]

Read the o/cers article. The S Code on the chip is from an OLD chip; a 2.0

Same markings here: http://www.hardocp.com/article.html?art=Nw==

From Ed:

" Update: There was a picture of the processor, it's just been yanked, but don't worry, I saved a copy.

It has an sspec of QFF8ES A4, and a type identifier of 80531.

Type identifer 80531 was used by Intel for non-Northwood 478-pin PIVs.

If you look carefully at the first picture in this article on the 2.0 Ghz PIV Willamette, you'll see that that CPU also has an identifying code of QFF8ES A4, and a type identifier of 80531 (though it's not the same processor as the one shown in the recent article).

If you don't want to look so far, here's a far clearer picture from [H]ardOCP (click on the first image). Again, QFF8ES A4, type identifier 80531.

Intel never recycles items like sspecs or type identifiers.

Occasionally, we've noticed websites put the wrong picture up for a processor, but the picture of the processor shows a code ending in 3.3G0K rather than 2.0G0K. "

Above quote from http://www.overclockers.com/tips00112/

There is little doubt that the image was photoshopped based upon that.

[bigben2k]

Ok, well, like I said, THG has an integrity issue, that doesn't change

The funny thing is that they posted a response to everyone's questions, and they said "We've done this before". That's kinda ironic...

Maybe it's just a sub-conscious attempt by Tom to kill his website, because he can't take it anymore... except that it's a conscious effort...

[pHaestus]

Well if you think about it as sort of a test for THG:

Post a fake picture, word everything in vague but misleading ways, and then see if anyone notices. If not then your job as a hardware reviewer just got a lot easier (takes less time to Photoshop than it does to actually review items).

But if people DO notice then you can make excuses and report to Tom that every little thing done on the site is closely watched and scrutinized by the world. This is a big ego stroker. Oh you also drive traffic up, eh?

Maybe I credit too much forethought to people, however.

[Heavy_Equipment]

Anyone notice the upper right corner of the Hardocp?

"Home of the real 3.3GHz CPU. (wink wink)"

The last time I went to THG, (prior to now) was on a link to the XP unlocking video. The reviews have always felt "off" to me.

Notice that there is no screenshots of any testing. Only their graph...which they input the data to. Ok, you can photochop the results photo too, but they can enter whatever scores they like into that graph.

I hope this just kills the site...I have no respect...or sympathy for any site that would pull a stunt like that.

Like someone else said, they just ruined their ties with Intel anyway. (Not to mention that of their source)