AMD Rolls Out new HD 7950 BIOS, Expects Existing Customers to Look the Other Way

[tbone8ty]

im liking these 7950 prices @ $299

http://www.newegg.com/Product/Produc...82E16814161407

[SKYMTL]

im liking these 7950 prices @ $299

http://www.newegg.com/Product/Produc...82E16814161407

At $299, I'd go for it in a heartbeat.

[Manicdan]

with how long these cards have been out, its only a natural progression im seeing.

[raghu78]

Honestly, now you are just trolling

I have no idea what you are talking about. BOTH the NVIDIA and AMD cards were left at their preset Power values. Setting it to anything else would be disingenuous and would not reflect the true state of the card. Having said that, by default, within the Boost BIOS, the AMD PowerTune limits are raised in order to take into account for the additional P-states. Without that, it wouldn't actually Boost to 925MHz. So the boost BIOS was used, AMD's stated speeds were achieved and the performance increased by the 5-8% every other site has achieved. I really don't know where you are pulling this "20%" number from.

In that same vein, I should have reviewed an overclocked, custom cooled GTX 660 Ti at launch. I didn't and quite conveniently, you seem to forget that. I reviewed (and BOUGHT) a card that was available at retail. I can't pick and choose what's available, nor will I intentionally stack the deck as you are stating. Sapphire's card may end up getting reviewed, maybe not but I won't be buying another, that's for sure.

Ok first up I will say I was wrong on the other partners not having HD 7950 reference cards. You were correct there. let me quote you.

"Setting it to anything else would be disingenuous and would not reflect the true state of the card."

So clearly you know that its getting throttled. Check out the xbitlabs review where they clock the sapphire dual x to 925 mhz.

http://www.xbitlabs.com/articles/gra..._11.html#sect0

The performance clearly scales by 12 - 15% wrt HD 7950 at 800 Mhz. Your review would have been much different with the power option at +20% . the HD 7950 would be closer to GTX 670 as it is in the xbitlabs review. But thats not something you want to show readers. clock for clock HD 7950 is 3 - 5% slower than HD 7970. Anybody who has seen other reviews knows that. Yeah you will buy a card to prove AMD's HD 7950 card is not good. Pick what you want on the GTX 660 Ti side. But why didn't you wait for a partner like sapphire to sample you the Sapphire HD 7950 Vapor-X. Do you buy all the cards that you review. That isn't possible. So why not wait for a good custom version to be sampled to you or wait and pick one yourself. You couldn't resist the temptation of having a go at AMD .
If you have really no bias pick the best clocking HD 7950 according to the reviews you have done and pit it against the best clocking GTX 670 in your reviews and let us all see what the results are. Until then you are trying hard to prove the HD 7950 is not as good as the GTX 670 which it is. And the HD 7950 is a better value than the GTX 670. The GTX 670 depending on sample can boost to 1.08 Ghz.

http://www.hardware.fr/articles/873-...asus-evga.html

Did you ever report the actual speeds the GTX 670 was running in your reviews. You won't because you don't want to show that GTX 670 at times is runnning at more than 250 Mhz higher clocks wrt HD 7950 at 800 Mhz. When you manually overclock the Kepler clock advantage vaporises and the true HD 7950 vs GTX 670 picture arises. Here is a clue about how to do it

http://www.hardocp.com/article/2012/...vs_670_hd_7950

the launch day review of Sapphire HD 7950 OC at HWC overclocked to 1200 Mhz. that a good start to pit against a custom GTX 670 OC. And don't forget to run the tests with power option at +20%.

http://www.hardwarecanucks.com/forum...oc-review.html

[DilTech]

You know Raghu... I have to ask you one simple question: How much stock do you own in AMD & Sapphire?

Seriously, all you've done is preached about that one specific card and you've ignored anything Sky has had to say. At least a reviewer has the balls to come here, give his notes on upcoming hardware (even when under NDA, tho he doesn't break it), and tries to fit in our requests into his already long and tedious review/article work! At the same time the guy doesn't fall for EITHER side's marketing BS that they spew to get reviewers to skew their results in either sides favor.

Consider this your one and only warning. This argument you keep trying to start ends NOW. One more post and that's that. It's been a long time since I've got to press the shiny red button, but I (and many others) feel that this has gone on for far too long. You're asking a guy who runs a review site to run out and spend a bunch of money just to shut you up when (and I know you won't believe me, but ANY reviewer will back this up) most review sites (except the HUGE ones) barely make enough from advertisement to pay their bills at the end of the month. Most only review what is sent to them, Sky ALREADY went out of pocket and bought a 7950 Boost Edition for everyone to see the numbers, now you expect him to spend another $300+?

KNOCK IT OFF!

As for you Sky, I tip my hat to your patience. I'm notorious for keeping a cool head, but I probably would've snapped on him already. Bravo.

[raghu78]

most review sites (except the HUGE ones) barely make enough from advertisement to pay their bills at the end of the month. Most only review what is sent to them, Sky ALREADY went out of pocket and bought a 7950 Boost Edition for everyone to see the numbers, now you expect him to spend another $300+?

This thread was just a result of over-reaction on the part of SKYMTL . He itself has accepted it

http://www.hardwarecanucks.com/news/...rns-unfounded/

As for custom HD 7950 Boost cards they are very good in acoustics / thermals .

http://hexus.net/tech/reviews/graphi...por-x/?page=12
http://www.kitguru.net/components/gr...ion-review/21/

I will not discuss this any further.

[mattlef]

This thread was just a result of over-reaction on the part of SKYMTL . He itself has accepted it

http://www.hardwarecanucks.com/news/...rns-unfounded/

As for custom HD 7950 Boost cards they are very good in acoustics / thermals .

http://hexus.net/tech/reviews/graphi...por-x/?page=12
http://www.kitguru.net/components/gr...ion-review/21/

I will not discuss this any further.

I think Raghu might believe if he pushes Sapphire enough, he might get a little mouth love from one of their booth girls at the next event he pays to go to.
Give it up dude. Sky has bettered your arguments time and time again. You and your jaded logic, while funny, is getting boring.

[SKYMTL]

Last post for me as well addressing some of these items. I won't ask questions since responses aren't necessary as I am merely educating anyone reading this post.

So clearly you know that its getting throttled. Check out the xbitlabs review where they clock the sapphire dual x to 925 mhz.

http://www.xbitlabs.com/articles/gra..._11.html#sect0

The performance clearly scales by 12 - 15% wrt HD 7950 at 800 Mhz. Your review would have been much different with the power option at +20% . the HD 7950 would be closer to GTX 670 as it is in the xbitlabs review. But thats not something you want to show readers. clock for clock HD 7950 is 3 - 5% slower than HD 7970. Anybody who has seen other reviews knows that. Yeah you will buy a card to prove AMD's HD 7950 card is not good. Pick what you want on the GTX 660 Ti side. But why didn't you wait for a partner like sapphire to sample you the Sapphire HD 7950 Vapor-X. Do you buy all the cards that you review. That isn't possible. So why not wait for a good custom version to be sampled to you or wait and pick one yourself. You couldn't resist the temptation of having a go at AMD .

I just want to point out that XBit Labs used a pre-overclocked HD 7950 WITHOUT PowerTune Boost. That can't be compared to a boost-enabled card. With overclocking, the non-Boost card will constantly stay at 925MHz. Meanwhile, the entire point of Boost is to hit 925MHz only when it is able to take advantage of TDB overhead. As such, OF COURSE the pre-overclocked card will be faster since the Boost card isn't hitting 925MHz all of the time, nor would it ever. NVIDIA has the same kind of affair with their GeForce Boost. It does tend to vary upwards and downwards much more than AMD's technology but that's besides the point since there is no way to compare Boost vs no Boost on their cards.

Regarding why I didn't wait; I was keeping a promise to you and many others. I promised to review a HD 7950 with Boost as soon as it became available. I didn't review a card that wasn't available, nor did I flash a BIOS onto a year old card to create some magical, never-to-be-available version as some did. Had I waited I'm sure some would have called bias there as well. Basically, I was pooched either way it seems.

If you have really no bias pick the best clocking HD 7950 according to the reviews you have done and pit it against the best clocking GTX 670 in your reviews and let us all see what the results are. Until then you are trying hard to prove the HD 7950 is not as good as the GTX 670 which it is. And the HD 7950 is a better value than the GTX 670. The GTX 670 depending on sample can boost to 1.08 Ghz.

Ironically, I never denied the HD 7950 being a better value above 1920 than what NVIDIA is offering, particularly now that we're seeing $299 cards.

Did you ever report the actual speeds the GTX 670 was running in your reviews. You won't because you don't want to show that GTX 670 at times is runnning at more than 250 Mhz higher clocks wrt HD 7950 at

Many seem to think that NVIDIA's Boost equates overclocking. It doesn't. NVIDIA's architecture is just meant to operate at higher clock speeds and has finer-grain control over clock vs TDP than AMD has at this point. Or at least that's the way it looks until AMD releases a tool that can actually track their Boost progress. So of course the GTX 670 is running at higher clock speeds; it was meant to do exactly that. We try to eliminate variance by running all cards in an environment that is temperature and humidity controlled so the playing field is quite level. Plus, now that AMD's cards have Boost, the argument is pretty much null and void anyways.

[Lanek]

This guy's getting destroyed but he keeps coming back for more

yeah, i have release my leather suit my whip and my handcuffs, i think i have find a good client there for it.

Many seem to think that NVIDIA's Boost equates overclocking. It doesn't. NVIDIA's architecture is just meant to operate at higher clock speeds and has finer-grain control over clock vs TDP than AMD has at this point. Or at least that's the way it looks until AMD releases a tool that can actually track their Boost progress. So of course the GTX 670 is running at higher clock speeds; it was meant to do exactly that. We try to eliminate variance by running all cards in an environment that is temperature and humidity controlled so the playing field is quite level. Plus, now that AMD's cards have Boost, the argument is pretty much null and void anyways

On this you are wrong, the AMD turbo boost is clock speed constrain too, there's no card who will get higher of the max who is fixed... it cant depass a certain speed.. + it is a lot more finer-grain controlled of what is doing the turbo boost of Nvidia... nvidia have only some notch at 13mhz, basically 4 differents clock speed are fixed.. the AMD one is doing it at 1mhz variation and is capable of change this a multiple time in 1 second,.. In reality, the AMD system was allready there, they have not get to change so much the system.. just to use it for a different approach ... Anandtech have a good report on it, i let you read it.

[SKYMTL]

On this you are wrong, the AMD turbo boost is clock speed constrain too, there's no card who will get higher of the max who is fixed... it cant depass a certain speed.. + it is a lot more finer-grain controlled of what is doing the turbo boost of Nvidia... nvidia have only some notch at 13mhz, basically 4 differents clock speed are fixed.. the AMD one is doing it at 1mhz variation and is capable of change this a multiple time in 1 second,.. In reality, the AMD system was allready there, they have not get to change so much the system.. just to use it for a different approach ... Anandtech have a good report on it, i let you read it.

Read the next line of the quote.

[BababooeyHTJ]

Many seem to think that NVIDIA's Boost equates overclocking. It doesn't. NVIDIA's architecture is just meant to operate at higher clock speeds and has finer-grain control over clock vs TDP than AMD has at this point. Or at least that's the way it looks until AMD releases a tool that can actually track their Boost progress. So of course the GTX 670 is running at higher clock speeds; it was meant to do exactly that. We try to eliminate variance by running all cards in an environment that is temperature and humidity controlled so the playing field is quite level. Plus, now that AMD's cards have Boost, the argument is pretty much null and void anyways.

I think that his main point is that most of the review samples being sent out seem to have a consistently high boost. I don't think that reviewers are receiving random cards like the rest of us. So compared to the cards that many of us may receive it is almost like a mild overclock. Its not the end of the world and its not enough to completely change results but still. Since clock speeds vary due to Nvidia's gpu boost knowing the clock speeds that the card actually runs at helps to know how other cards stack up to it.

[SKYMTL]

Interesting assertion. I can confirm that every one of my cards comes in retail packaging and the BIOS version lines up perfectly with the retail BIOS version. In addition, they are typically sealed in plastic from the factory, ready for retail channels. Due to the nature of releases, PR guys wouldn't even have the time to unpackage cards, bin and then send out to reviewers in time for launch as usually they only get the cards from the factory hours before having to ship them out.

The only exception to that was a recent ASUS card that was supposed to be sent to a system integrator's QA lab and landed at my doorstep. That was rectified in VERY short order after ASUS realized their mistake.

[Lanek]

Interesting assertion. I can confirm that every one of my cards comes in retail packaging and the BIOS version lines up perfectly with the retail BIOS version. In addition, they are typically sealed in plastic from the factory, ready for retail channels. Due to the nature of releases, PR guys wouldn't even have the time to unpackage cards, bin and then send out to reviewers in time for launch as usually they only get the cards from the factory hours before having to ship them out.

The only exception to that was a recent ASUS card that was supposed to be sent to a system integrator's QA lab and landed at my doorstep. That was rectified in VERY short order after ASUS realized their mistake.

Personally if i was send you a card, i will choose the card first.. but thats only me.