AMD Radeon HD 6950 Can Be Unlocked To HD 6970

An anonymous reader writes "AMD's new Radeon HD 6950 can be unlocked to a HD 6970 via BIOS mod. Performance of the unlocked card is identical to the full blown HD 6970!"

Comments

It is still different HW

[BadAnalogyGuy]

A lot of manufacturers will do this, actually. Their first device will contain very high quality, standard HW that is somewhat overspec for what they intend, but due to driver support and ease of implementation they can get it out the door in a reasonable amount of time. Then for their successor device they will take the lessons learned, use cheaper parts, use better optimized software, and sell it as the "cheaper" version.

You are getting lousier HW, but arguably better SW, so the performance gap isn't as big as their marketing lit will let on. On paper, the expensive first gen device looks better, but when the rubber hits the anus it's pretty much a wash.

Re:It is still different HW

[ZDRuX]

"...but when the rubber hits the anus it's pretty much a wash."

I was like whaa....? But then I looked at your name and all was set right in the universe.

Re:It is still different HW

[cheater512]

Considering I've got a AMD X3 and Linux is reporting 4 cores, I'd agree with you.

They do bin the chips based on performance, but that doesnt guarantee that you dont get a higher performance chip, nor does it mean you cannot use the additional hardware assuming its not completely dead.
In my example, I've obviously gotten a chip that was binned as a AMD X4 but they needed more X3's so they just turned a working core off.

Re:It is still different HW

[SuricouRaven]

Or perhaps they turned it off because, while it works almost all the time, it'll fail one in ten million floating point operations at random, or is prone to fail at moderatly high temperatures or workloads. If you want to use the 'disabled' core, I suggest you run your own tests to determine if there is some minor fault. Slow the fans so it runs hot and calculate pi. If it can run for 24 hours and produce the right result, it's probably good.

Re:It is still different HW

[cheater512]

Its been running for 18 months. I'm quite confident what I got was indeed a X4 or at least any problem with it is so minor it can not occur with my usage patterns. :)

Re:It is still different HW

[drinkypoo]

I ran for maybe a year overclocked and then one day while playing Alpha Centauri and watching a video at the same time my system became unstable... back to stock speed, no more problems. Good luck! Silicon DOES degrade over time.

Re:It is still different HW

[realityimpaired]

That actually does make sense from a fabricating standpoint... crippleware, that is. It's a lot cheaper to just have one product line to manufacture, which would allow them to take advantage of economies of scale. They can then split it into two product lines by disabling certain features, or lowering some of the spec in software. I'd wager it's a *lot* cheaper than building/operating a separate fab, and separate supply line for the other product line. :)

They wouldn't be the first company to do that, not by any stretch of the imagination. It's actually pretty common in home electronics... Who here doesn't know somebody who's got some stereo or DVD player that has a USB port it can't use? Or a camera that looks identical to the top-of-the-line version, but is lacking that fine-grained white balance tuning?

Re:It is still different HW

[Lord Byron II]

Just to reply to the parents: -Just because you run the system for 6/12/18 months without a crash does not mean its stable. Most apps won't crash if 1+1=1.9999. You're not likely to notice if a single pixel in a single frame in a YouTube video is the wrong color. -Just because your system is stable, then has a crash doesn't mean that the silicon is degrading. Even a "perfect" chip will have a fault or two every few 10^x calculations (where x is some large number).

Re:It is still different HW

[Kreigaffe]

An image of Steve Jobs hovering precariously atop the production line, straddling the birth tube of his empire, a smile on his face as he hunkers down and lets the conveyor belt drag each and every item ever so gently under his balls for the Reality-Distortion-Teabag.

Re:It is still different HW

[hairyfeet]

But how stable will it be when the HD6970 uses both an extra 50w AND a different pin connection than a HD6950. The 6970 uses an 8 pin and a 6, the HD6950 uses a dual 6 connector setup. Can dual PCIe connectors pull the 250w load that the HD6970 requires? This isn't like some Athlon X3 where the only thing you have to do is flip a switch, the other chip pulls more wattage at load and there is no telling if there are other parts besides the connector pin set that they have changed between the cards. Since both cards are over $300 and I'm sure the BIOS flash kills the warranty that is sure a lot of "ifs" there.

Re:It is still different HW

[Fulcrum of Evil]

I have an iPhone and it's the first phone that doesn't break in stupid ways that piss me off. What can I say, I like having something that actually works.

Re:It is still different HW

[cheater512]

I dont know how overclocking got in to this conversation. In my instance there is no overclocking - just unlocking a extra core that works correctly.

In the article's instance, its unlocking the additional pipelines and so on, with optional overclocking.

If this

[velja27]

If people start to buy this kind of "locked" graphic cards and unlock them then the manufacturers will start to cripple the cards for good. Or simply make truly weaker graphic cards instead of limited ones with the same chipset.

Re:If this

[hellop2]

So, what's your point? Don't hack hardware?

We can't control what they do. Luckily, they can't control what we do.

Re:If this

[larppaxyz]

Long time ago, i was using a tool that let's you modify any NVidia card BIOS. I very soon had dirty words appearing when BIOS POST was running. It also made it possible to change stock GPU and memory clock frequencies.

Re:If this

[Zakabog]

Luckily, they can't control what we do.

That's his point, they can control what we do. If we hack their hardware to run better with simple software solutions then they'll just redesign the hardware so there's a physical restriction on how well the card will perform. Though there would be no point in being able to hack the device if you're too afraid to do it for fear that they might cripple future devices.

Re:If this

[antifoidulus]

Ultimately it depends on how much they actually (think) they end up losing. The reason they ship the same hardware to begin with is because it obviously costs them more to set up separate lines for the two cards than it does for them to put the extra bit unused hardware in the lower spec'd card. If ultimately they think that doing this will cost them more in lost sales than they gain in increased manufacturing efficiency then they will start shipping divergent hardware, but if only a tiny portion of their customer base mods their cards then they will probably just consider it collateral damage and maybe up the difficulty of modding the software in the next version.

Re:If this

[Rockoon]

That's his point, they can control what we do. If we hack their hardware to run better with simple software solutions then they'll just redesign the hardware so there's a physical restriction on how well the card will perform.

Why presume that these hacks arent just a profit-motived feature?

What incentive do they have not to do things like they are now? or more to the point, if they did "permanent" disabling, what would be their incentive not to change to a disable method with an easy workaround?

I think that you people are forgetting your history. Both AMD and Intel used to always do real physical disabling. Now they don't always do that.

O! the Humanity!

[jabberw0k]

If people start buying underclocked CPUs and overclocking them, the manufacturers will start to cripple the CPUs for good, or make weaker CPUs... Wait, haven't we been down that road before?

Overclocking guide

[nicholas22]

An overclocking guide can be found here. You *might* get problems under extreme load, because the 6950 uses the 6-pin power connector, whereas the 6970 can draw more power, because it uses the 9-pin connector.

Re:Overclocking guide

[Elledan]

That's not correct. The 6 and 8-pin PCIe connectors are identical. They have the same number of ground and 12V wires between the GPU and the PSU, the same wire gauge and can carry the same amount of power. The 8-pin connector exists because in the PCIe spec they had a sense wire for the 12V line specified on this connector, which would then allow the connector to carry more current as the PSU would be able to better regulate the voltage. In practice this is much more easily done at the PSU side, making the 8-pin connector useless and allows the 6-pin connector to carry the same 150 Watt as the 8-pin one.

Want to check this? Just use a 6-pin connector and short the remaining two pins on the GPU to ground to satisfy the GPU if it checks for a connection there and everything will work just peachy fine. If you check 8-pin PCIe connectors you'll see that this is all they do: short the two extra pins to ground.

9500/9700 all over again

[lostmongoose]

same thing was possible with the 9550 Pro -> 9700 and with the 9700 -> 9700 Pro both were done with BIOS flashing

6950 -- 6970

[dangitman]

So, that's 20 faster?

Re:6950 -- 6970

[TeknoHog]

It's only 0.29 % faster, probably not worth it.

3.5" floppies

[bonniot]

Reminds me how the way drives recognized 1.44MB floppies (3.5") from 720KB ones was by checking if there was a hole in the bottom-right corner (the bottom-left corner being for write protection). And sure enough, if you made a hole in a 720KB floppy it would be possible to format it as 1.44. There might have been a few more errors, but I remember when HD floppies were 3-4 times more expensive, so it was definitely worth it. At least for a teenager with only pocket money. Ah, those floppy drilling afternoons... Mais où sont les neiges d'antan?

bad for the environment

[Anonymous Coward]

Welcome to the world of "economy of scale". You may also be interested to see how routers are sold/marketed. (see: dlink, linksys and motorola, etc)

The interesting part is NOT why this can be, but rather whether it is legal and also how the bios code was extracted. This is why consumers should demand more "open hardware". Because consumers are consistently paying for the manufacturing of quality hardware only to have the manufacturer bundle crap software (cripple ware) onto it. And for what? so they can target various price points within their target markets.

Question: do you think this is an environmentally sound practice? It isn't very "green" to sell a physical product to the consumer only to restrict its usage to some lesser subset of its full potential. I don't understand why the geeky tree huggers among us don't get on this and start demanding more and longer functionality out of the products we consume. ex: I have access to 4 cannon cameras that look much the same, are the same age and yet all the batteries and chargers are incompatible. Why? so i cannot reuse batteries or charges and must trash and re-buy each individually. With other products I cannot charge everything thru USB even when it is possible. Why? for additional after market adapter sales. I have routers that are exact same hardware yet function and priced quite differently. why? do i have to tell you. I have portable devices that have rechargable AA batteries taped into a pack with a unique plug, instead of just using a regular AA slot. Why? i think you know. Cell phone pricing/plans/contracts/packages are designed to encourage me to "revolve" phones every year or so... why? its too obvious to say.

We consumers are being forced to make additional garbage for the landfills and discouraged from thinking about the consequences when we really could squeeze much more life from our existing electronics. We should be outraged (those that care for the future). But instead we are lulled into the belief that our existing equipment is crap and that getting something new benefits us. We are convinced that we are the ones demanding this from the manufactures. I tell you it is the other way. In reality this only benefits the manufacturer... who is actively limiting the functionality of our beloved products to further this fallacy to maximise their profits.

What you can do when possible (if you care): Buy generic brand electronics, use open source and demand refunds for bundled software when possible. Note that Windows is always refundable when sold with a computer... read the contract... if you care to read contracts before accepting them. ie: keep the quality computer hardware, but drop the cripple wear (windows 7 starter).d

Re:bad for the environment

[Anonymous Coward]

I don't think "doing what I want with stuff I bought and own" to be much of an entitlement, personally. :-P

Re:bad for the environment

[Kjella]

And for what? so they can target various price points within their target markets.

Well, let's say you want to replace the 6950/70 with a single product, what should the price be? They can make it as expensive as the 6950, losing a bunch of profit on the 6970 sales they would have made. Or as expensive as the 6970, but then you'd lose lots of sales that just wouldn't pay what a 6970 costs. You talk as if price points were made up by the industry when they in fact reflect that people have different amounts of disposable cash and put different value on things. There's not a single card that would simultaniously satisfy both markets and be a sound investment for AMD. And if serving each market with its own design costs more, then of course they choose a common design as anything else would be just stupid.

Maybe it's easier to see the point with software and features than hardware and binning. The difference between Photoshop and Photoshop Elements is probably a #ifdef in the source code. Why then should they offer a "crippled" version? Well because some people are interested in paying for the missing features and some aren't. It's reasonable that the people that want CMYK separation pay for CMYK separation and the rest don't need to subsidize them. Those who pay for the 6970 are the ones most willing to pay for performance and so they do. Isn't it completely reasonable and logical they cover more of the R&D tham the 6850 owners?

Question: do you think this is an environmentally sound practice? It isn't very "green" to sell a physical product to the consumer only to restrict its usage to some lesser subset of its full potential.

Actually, the costs in developing two product lines are probably larger both environmentally and economically. That it would be offset by a better design is unlikely, since the difference between the two boards is largely making use of more energy-efficient but slower components. If there was a significant saving to die size, they would do it quite quickly as on this point since the economic and environmental cost are largely aligned.

It's about manufacturing yield...

[EmagGeek]

So if I am a graphics chip manufacturer, I know that the fewer unique designs I have, the cheaper it will be to manufacture my product line. If I make both chips and boards, the same economy of scale applies to both the chips themselves and the assembled boards.

If I can determine both my chip and board yield at in-circuit test, and configure each manufactured device to its maximum possible stable capability, then my manufacturing product yield is maximized.

This type of yield binning is nothing new.

Re:It's about manufacturing yield...

[Anonymous Coward]

Except you're wrong about how parts are binned. It's generally not so much about yield binning, and more about market segments, 10% of the market will buy the best part for twice the price, 80% will buy whatever is best value for money, and the last 10% will buy whatever is cheapest (percentages made up), the trouble is when you are in the business of semiconductor manufacture, and you have tweaked your process to 99.98% yield with 0.02% tolerances, over 80% of the parts coming off the line are topend parts, just about your entire cost of manufacture is in plant capital, and 90% of your profit comes from the top 10% units. You basically have to create an artificial lowend to create a highend market. Generally it makes more sense to fuse parts to lower spec, as softmods basically ruin sales of topend devices.

Environmentally friendly, fuck no. Required business practise, almost certainly.

It also doesn't make a lot of environmental sense to mass-produce parts which are obsolete in 6 months, it would be better from that perspective to develop semiconductor fabs until they reached their natural atomic-resolution limits, and then produce parts that are as good as they practically can be. But that would again completely canabalise their business and they'd go bust. It would also be a disaster for everyone who needs more computing power right now, and the results of that would probably be much worse than the silicon waste we have now (which is mostly energy and gold, as silicon, copper and boron are abundant and cheap).

5.25" floppies: instant 2x space

[whovian]

It was possible to use the flip side of 5.25" floppies by notching the other edge of the disk. Specialized cutters were sold for making square notches, but round-hole paper punchers worked too. Manufacturers certified, of course, only the original side of the disk, but I never had a problem using the flip side.

Thanks for the link - psst, don' tell anyone else

[G3ckoG33k]

Thanks for the link - psst, don' tell anyone else or else AMD will stop it

Re:Nothing to see here, move along

[JustOK]

Some people just don't listen during history class, that's all.

That's certainly something we've learned from history...over and over and over again.

3 words for you

[thegarbz]

I got 3 words for you. Economies of scale.

It underpins the manufacturing of all processors currently out there. What is being done here is nothing new. It's as old as overclocking. Some manufactures tried various ways of locking them. Ultimately though once you customise a chip enough it adds to the bottom line production cost. That's why AMD's version of hardware locking at the time included setting 5 jumpers external to the CPU die and then laser cutting them. Remember the pencil trick to get Athlons unlocked?

Creating a truly weaker card means customised production runs, which means setup costs for the batch. Not something you want when margins are next to nothing.

Re:3 words for you

[DaveGod]

TFA is reporting a 100% success rate, but don't assume every 6950 is going to mod as well as this batch. It may be that the foundry have utterly nailed this chip and every single one is going to roll out at 6970 spec. But, it's probably just scale economies on the initial supply, scale economies do change over time. It might also be that *spins wheel* they deliberately did this because the 6970 is the only fully tested board yet. Or *spins wheel* they wanted these rumours to help sell their cards.

Or *spins wheel* their BIOS is coping brilliantly with potential problems, so even failures are being perceived as successes. For example, AMD's new approach is to underclock cards when they get near the TDP limit. Perhaps a "genuine" 6970 is only ever downclocking itself on Furmark, while the modded 6950's are doing so much more easily - quite plausable given you still only have the 6950 power connectors.

Anyone planning on buying one of these cards just found another reason to grab one sooner rather than later, but personally I'm going to be waiting a while anyway before considering modding mine.

Re: Crippleware

[Kreigaffe]

nah you're looking at it wrong.

graphics card makers have been doing this for many many years.

so you get a new card design. the biggest cost is going to be setting up production for it. once everything's all tooled up and ready to go, your biggest expense is out of the way.

then you get a better design. you tool up for it, but you still want to sell the older design at a lower cost -- not everyone wants to drop bank for top of the line. except, dang. you're not making those cards any more. well, rather than go through the expense of re-tooling everything and doing limited runs of an inferior product... it's a much better decision to just take what you are producing and cripple a certain number to meet the specs for the old card and sell them. Sure, profit margins might be more slim than had you stuck with the original tooling, but it's important for your production to be of the most advanced card possible -- so you won't need to redesign and retool everything for the longest time possible.

basically, it's probably quite a bit cheaper for them to sell a higher-tier card for a lower price with some features disabled than it would be for them to take production offline and retool for an already-obsolete design

We used to do that

[0123456]

When I worked for a hardware company, we built one basic chip and different boards used a different BIOS. In theory you could have installed a BIOS from the fastest cards --- if we hadn't blown fuses in the chip to prevent that from working -- but the high performance boards used the chips which had been proven to work reliably at those clock speeds with all components enabled, while the lower performance boards either didn't check out at the highest clock speeds or didn't pass all the hardware tests so some parts of the chip were suspect and had to be disabled.

So in our case if anyone had managed to get a different BIOS working on the chip and work around the blown fuses, odds are it would be flakey or simply refuse to work.