MSI and ASUS Send VGA Review Samples with Higher Clocks than Retail Cards

[moproblems99]

Just figured I would chime in and say that you are assuming people only care about the stock performance of graphics cards. That carries about as much weight with most people as the stock performance of a car, truck, or bike. You are correct in that the average user don't understand the capabilities of cards. But those same users also don't get online and read reviews either. They run down to the local shop and ask what should I buy? They will get told the trendy answer.

 

[Frick]

Just figured I would chime in and say that you are assuming people only care about the stock performance of graphics cards. That carries about as much weight with most people as the stock performance of a car, truck, or bike. You are correct in that the average user don't understand the capabilities of cards. But those same users also don't get online and read reviews either. They run down to the local shop and ask what should I buy? They will get told the trendy answer.

Pretty sure this isn't true.

 

[najiro]

True, not ALL buyers read reviews or do elaborate research before buying a $400+ hardware but I believe a larger percentage of buyers today are wiser and have access to the internet. Even if its just Amazon and they read the product feedback there, it counts. Note that some stores also put up the product's awards from reviews like Newegg. Still, the idea that end-users do not care or are okay with this kind of elaborate scheme is something I'm really against. It would likely boost these brands' confidence about such and get more 'creative' in cheating in the future.. overall, not good..

 

[cadaveca]

Pretty sure this isn't true.

Yeah, they are just as likely to watch a video on YouTube. Component reviews cater to enthusiasts only, who, for the most part, don't even care about stock performance; they are looking for the card that either runs the coolest(quietest) or the one with the beefy VRM for OC.

True, not ALL buyers read reviews or do elaborate research before buying a $400+ hardware but I believe a larger percentage of buyers today are wiser and have access to the internet. Even if its just Amazon and they read the product feedback there, it counts. Note that some stores also put up the product's awards from reviews like Newegg. Still, the idea that end-users do not care or are okay with this kind of elaborate scheme is something I'm really against. It would likely boost these brands' confidence about such and get more 'creative' in cheating in the future.. overall, not good..

I'd agree with you if there was a truly significant difference in performance offered by 30 MHz on a GPU that is over 1500 MHz, but the fact remains that that 30 MHz = <1FPS in nearly every instance.

You'll also note that W1zzard lists clocks in his reviews on the first page, and the review lists the "OC" clocks for both the cards in question. He didn't copy numbers from marketing material; he likely tested the card with his own tool, GPU-Z, that he wrote, and got the clocks from there.

Apparently you are CLUELESS about review standards.

Um, I hate to break it to you, but there is no such thing in the real world as review standards. I do motherboard, memory, mini-PC and other reviews for this very site. Our "standards", if you want to call it that, are dictated by us, and by W1zzard. There is no REIVEW STANDARDS BOARD or REVIEW ETHICS COMMITEE that all reviewers are a part of. I am pretty much free to do whatever I like in my own reviews, as long as I stick to FACTS.

FACT: The GPUs in question were tested as they were provided.

FACT: The review states the clocks the cards were tested at

FACT: The review tested the card at those clocks.

FACT: When it was discovered that the tested clocks were not the same as retail, something was done about it.

FACT: Each of these GPUs is "verified" by the OEM to run these clocks if using a tool, and just because some users do not like the tool, doesn't mean the review is posting wrong information. Software tools at this day an age do not affect performance.

FACT: Amazon and Newegg reviews are posted by general users, not by "reviewers", and tend to have more favoritism in them and BS than I am comfortable with.

 

[EarthDog]

Well said!

I am going to guess we get the cards in "OC Mode" as those two have apps with higher clocked profiles. I don't believe EVGA does have the little penny app that does that for you with one button. I know, like TPU (which huge +1 to Dave's post above), they mention what the clocks are. I know I have mentioned a card comes in overclocked mode and list the ACTUAL (like from a real time monitor that shows true boost) clocks on top of it.

I can't say I feel deceived. Perhaps more aware than some others maybe?

 

[MxPhenom 216]

Yeah, they are just as likely to watch a video on YouTube. Component reviews cater to enthusiasts only, who, for the most part, don't even care about stock performance; they are looking for the card that either runs the coolest(quietest) or the one with the beefy VRM for OC.



I'd agree with you if there was a truly significant difference in performance offered by 30 MHz on a GPU that is over 1500 MHz, but the fact remains that that 30 MHz = <1FPS in nearly every instance.

You'll also note that W1zzard lists clocks in his reviews on the first page, and the review lists the "OC" clocks for both the cards in question. He didn't copy numbers from marketing material; he likely tested the card with his own tool, GPU-Z, that he wrote, and got the clocks from there.



Um, I hate to break it to you, but there is no such thing in the real world as review standards. I do motherboard, memory, mini-PC and other reviews for this very site. Our "standards", if you want to call it that, are dictated by us, and by W1zzard. There is no REIVEW STANDARDS BOARD or REVIEW ETHICS COMMITEE that all reviewers are a part of. I am pretty much free to do whatever I like in my own reviews, as long as I stick to FACTS.

FACT: The GPUs in question were tested as they were provided.

FACT: The review states the clocks the cards were tested at

FACT: The review tested the card at those clocks.

FACT: When it was discovered that the tested clocks were not the same as retail, something was done about it.

FACT: Each of these GPUs is "verified" by the OEM to run these clocks if using a tool, and just because some users do not like the tool, doesn't mean the review is posting wrong information. Software tools at this day an age do not affect performance.

FACT: Amazon and Newegg reviews are posted by general users, not by "reviewers", and tend to have more favoritism in them and BS than I am comfortable with.

I was going to say something about "review standards" but since i dont do reviews i wasnt sure.

 

[Chaitanya]

Here is Msi's official statement regarding accusitions:
http://www.overclock3d.net/articles...ment_regarding_oc-mode_on_msi_gaming_x_gpus/1

 

[Vayra86]

Well said!

I am going to guess we get the cards in "OC Mode" as those two have apps with higher clocked profiles. I don't believe EVGA does have the little penny app that does that for you with one button. I know, like TPU (which huge +1 to Dave's post above), they mention what the clocks are. I know I have mentioned a card comes in overclocked mode and list the ACTUAL (like from a real time monitor that shows true boost) clocks on top of it.

I can't say I feel deceived. Perhaps more aware than some others maybe?

I don't feel deceived either, but that is because we are visitors of a tech forum and knee deep in tech mumbo jumbo, closely following these releases and have knowledge of how a GPU actually works.

There is a difference here. Reviews are being read not just by nerds , but by all consumers. I still think it is a rather shady business, just like motherboards with a 1% clock bump.

 

[cadaveca]

I don't feel deceived either, but that is because we are visitors of a tech forum and knee deep in tech mumbo jumbo, closely following these releases and have knowledge of how a GPU actually works.

There is a difference here. Reviews are being read not just by nerds , but by all consumers. I still think it is a rather shady business, just like motherboards with a 1% clock bump.

But if such info is contained within the review, where's the problem?

Since I do the motherboard reviews here. you'll note I myself include CPU-Z screenshots for expressly that purpose (as do most other websites).

I could do a review with absolutely ZERO benchmarks, and still get my point across, but everyone wants benchmarks, so they get included. If "normal" users are reading reviews, they should also notice things like BCLK speeds, and things such as what this thread is about, too.

There is also advance Turbo profiling in some board's BIOS that affect performance as well. I do make a point of mentioning it in my reviews if it is present. I hounded ASUS about this in my reviews, and today if they use an advanced Turbo profile (such as when XMP is enabled), they give a pop-up in BIOS asking if you would like to enable it.

To me, it seems as most users complaining about this are quibbling about minor details, however, thanks to this, MSI is releasing the same BIOS for the GPU in question to the general public, or you can get these BIOSes here on TPU in our BIOS database.

 

[wellnow1101]

I personally find this hyped up BS to nothing at all. If you want the performance you have to OC the damned thing. Was this used as an excuse to increase site visits? Can't be sure but, imho .... this is nothing but hype for readership.

 

[cadaveca]

I personally find this hyped up BS to nothing at all. If you want the performance you have to OC the damned thing. Was this used as an excuse to increase site visits? Can't be sure but, imho .... this is nothing but hype for readership.

Not a thing to generate traffic. We have plenty.

The issue is that these companies advertise these speeds, and do support them via software in retail units, but the units provided for reviews do not require the software. That makes the reviewed item and the retail item different, which may miss-inform the end user reading a review.

In the end, the differences in performance are small though.

MSI has gone as far as responding, and then releasing the BIOS on the review cards to the general public, which is a fantastic move by them. With that done, users that do not want to run any software for OC have a supported BIOS that can be run, and do not have to "OC", as you put it.

Wanting what the end user gets to be exactly the same as what is sent to reviewers is worth taking action.

 

[rtwjunkie]

I personally find this hyped up BS to nothing at all. If you want the performance you have to OC the damned thing. Was this used as an excuse to increase site visits? Can't be sure but, imho .... this is nothing but hype for readership.

I'm not on the "burn down MSI and Asus bandwagon", nor am I excusing what they did.

It is however, real, not hype. It has been admitted. It was something two separate sites noticed and investigated. If anything, it has strengthened the credibility of TPU for not being complicit in the deception.

 

[Ubersonic]

I don't feel deceived either

I do, I bought my MSI Gaming 980ti based on the performance/noise stats I read on TPU, and today I found out that TPU were sent a clocked up review sample. Not only have I been cheated out of performance for a year by MSI but now I'm going to have to scour the internet to try and find a dump of their review BIOS so I can get the performance/noise stats I thought I paid for 12 months ago.

 

[rtwjunkie]

I do, I bought my MSI Gaming 980ti based on the performance/noise stats I read on TPU, and today I found out that TPU were sent a clocked up review sample. Not only have I been cheated out of performance for a year by MSI but now I'm going to have to scour the internet to try and find a dump of their review BIOS so I can get the performance/noise stats I thought I paid for 12 months ago.

Wow, I wish I had your real world problems, where that is the biggest issue in my life...

 

[Ubersonic]

Wow, I wish I had your real world problems, where that is the biggest issue in my life...

If your MSI 660 was a Gaming not a TF then you would have this problem, that's how long they've been doing it

 

[Ungari]

It seems to me that more people are upset with the side issue of software controlled settings, than the original topic.
Had as in other cards this OC Mode Setting been activated by flipping a toggle switch on this card, would we even be talking about this?

 

[cadaveca]

It seems to me that more people are upset with the side issue of software controlled settings, than the original topic.
Had as in other cards this OC Mode Setting been activated by flipping a toggle switch on this card, would we even be talking about this?

No.

Flashing BIOS to a card can go wrong in so many ways, so easily in comparison to flipping a switch.

What makes it even more weird is that it is done for such small increases.

 

[Ungari]

I must say that I prefer the older method of a physical BIOS switch than using the user defined software which I think is now prevalent due to the RGB lighting trend.

The difference between these modes is for the ability to tout it's performance per watt efficiency. Had it been stated more clearly that this spec was solely based on the mid-range default BIOS setting, perhaps the retail cards could be shipped with the higher OC out of the box without conflict.

 

[cadaveca]

I must say that I prefer the older method of a physical BIOS switch than using the user defined software which I think is now prevalent due to the RGB lighting trend.

The difference between these modes is for the ability to tout it's performance per watt efficiency. Had it been stated more clearly that this spec was solely based on the mid-range default BIOS setting, perhaps the retail cards could be shipped with the higher OC out of the box without conflict.

More than likely.

Personally, I'm just left confused by the whole thing. Both MSI and ASUS have great cards here. They don't need to do things like this to have their cards look great in reviews. MSI has a nearly silent cooling system, with aesthetics that are all their own.

ASUS has their own style, with built-in non-disclosed features for OC. Completely different product focus between the two of them. The actual clockspeeds of the GPUs aren't that important when the differences are so small, yet design focus is so different.

There used to be a day where we could expect a 10% overhead in GPU clockspeed, attainable by OC. Having a BIOS that boosts GPU speed a bit reduces that percentage overall.

 

[Ubersonic]

It seems to me that more people are upset with the side issue of software controlled settings, than the original topic.
Had as in other cards this OC Mode Setting been activated by flipping a toggle switch on this card, would we even be talking about this?

No because we would have received the same card as reviewers did (the one we thought we paid for), the reviewers would have shown different scores for the turbo switch on/off and we would have known to turn it on.

There was no warning from MSI that retail out of the box performance would be inferior to review sample out of the box performance. That's what has annoyed many of us.

 

[Ungari]

No because we would have received the same card as reviewers did (the one we thought we paid for), the reviewers would have shown different scores for the turbo switch on/off and we would have known to turn it on.

There was no warning from MSI that retail out of the box performance would be inferior to review sample out of the box performance. That's what has annoyed many of us.

Had you been told that the default BIOS for consumers was the middle preset, with the ability for consumers to select the exact same OC preset using their software, would it have been better?

 

[Ubersonic]

Had you been told that the default BIOS for consumers was the middle preset, with the ability for consumers to select the exact same OC preset using their software, would it have been better?

Yes because I would have known that I needed to either source a copy of the review card's BIOS and flash mine or just buy another card.

NB: For anyone with a retail MSI Gaming 980ti the "real" BIOS can be downloaded here:

https://www.techpowerup.com/vgabios/173582/msi-gtx980ti-6144-150622

 

[EarthDog]

Yeah, those 38 MHz REALLY make a difference! 24 MHz on the 1070! Makes a game playable or not, right? Can go from high to ultra with that 1-2% performamce difference in OC mode.

 

[bug]

Yeah, those 38 MHz REALLY make a difference! 24 MHz on the 1070! Makes a game playable or not, right? Can go from high to ultra with that 1-2% performamce difference in OC mode.

I think I've posted this before, but I'll post it again.
That kind of difference only matters in reviews. Because all 1070 customs designs will be virtually identical, that 1-2% can make one card (artificially) appear a hair faster than another. This can sway one's buying decision. Personally, I think it's more insulting having 3 supposedly "overclocking" profiles, all within 100MHz.

 

[EarthDog]

It's not artificial though. Every single card can hit those speeds with a touch of a button. The difference here is only in touching a button or not.