Slashdot Mirror
AMD Downgrades Bulldozer Transistor Count By 800 Million
Robadob writes "It has come to light that AMD PR had originally reported that the new Bulldozer processor's transistor count was 2 billion. AMD PR are now asking reviewers to correct this count to 1.2 billion from the original amount they provided ~3 months ago."
Backwards into the future.
To offset political mods, replace Flamebait with Insightful.
NO, you stupid AMD, don't do that...
The software retards get away with worse all the time, and *still* blame the hardware when their software is slow/buggy.
Get rid of 800 million transistors? Sure 800,000,000 vacuum tubes it is.
but this 66% is really outrageous
The FPU in these chips rounds 1.2 billion up to 2.0 billion.
With the new ceo ?
Normally the route of a true american corporate cultured corporation would be to deny everything and fool everyone and rip as much cash as it can. Until they were confronted at courts.
But now, amd marketing is rather needlessly contacting reviewers to make corrections, while taking a hit in p.r.
But is it really a hit ? Coupled with the fact that the new ceo kicked a lot of marketing staff, this tells me that the new term in amd is going to be a term reminiscent of early 90s in technology - a responsible era in which corporations have actually manufactured useful gadgets and sold them honestly, trying to get the edge on each other through tech - not with filthy dealings or deceit (hello intel and the bribery verdict)
Read radical news here
Seems the outsourced replacements punched in the "round up" function on the Excel spreadsheet instead of "round to nearest"...
I'm paying for *transistor count*, not GFLOPS!!!
Priorities, people!
I understand the importance of truth in advertising, but is this information meaningful, or just an insignificant correction? The magnitude of the difference alone doesn't automatically make this an important story, or the exposure of some big, inexcusable lie by AMD.
What's the true relevance of transistor count? If I see two processors with identical performance and power efficiency but radically different transistor counts do I have any real world incentive to select one over the other? I mean, presumably the one with fewer transistors in roughly the same die space might overclock better, might have a longer MTBF, etc., but beyond that should I care?
Or did timothy post this just to keep up the fanboi flame wars?
One has to wonder why they were using Pentium processors to calculate the transistor total, though.
I guess the new figures make a little more sense. Bulldozer's performance was fairly similar to their previous (and smaller) Thuban Core, at 904 million transistors -- it was as if AMD decided to take more than half of their transistor design budget, heap it in a corner, and set it on fire.
Does anyone have any idea why and how this happened? I mean, could there have been an intention to defraud or is this some sort of epic fail? TFA doesn't give much information.
It's not like we are going to count this any faster.
WTF: this thing is fabricated in America?
PS: I was at an Irish Fair, so there were plenty of Duhloreans around for me to test my Paradox theories on the Hewie Lewiss fan-base.
PS2: Most chips are fabricated in non-American countries, so this would lead me to suspect that one-day America will return to being a continental free frontier country as it was back when it was founded on Alcohol and Tobacco and Firearms and Cotton with a new addition of Silicon fabrication. BATFE is un-American in this regards, and America has always been about displaced tribes and convicts being exported to re-build their character and fortune in America cheaper than the rest of the world and it's always been the previous 50 years of China's booming economy that stole the Light of Liberty & Freedom from Americans but that will change soon when Obama is re-elected.
PS3: I just got back from the future, and everyone is sure that they'll reclaim their freedoms on a 5th election of Obama and Palin this time as his running mate. The fire-side chats of Obama and Palin show that muslim-christianty and communism can work together like it has done in east-Africa and that we should continue the legacy of exporting jobs to foreign countries and preventing the world from spending US currency in America because like it was in 2012 just because illegal aliens have 9 of 10 US domestic dollars doesn't mean they have a right to come to America to spend their US currency and likewise it's no different even thought they have 99.9% of US currency more-so than 90% in prior years.
PS4: The Amish are running under-ground milking farms that are interfering with everyone's subcutaneous pharmaceutical dispensery rations and cerebral bio-meters, and so we need to change the early founding-father's policy of religous freedom as excluding Amish apothecaries and Amish farming.
So a few points about this rather bizarre announcement:
1. Unfortunately for AMD this does nothing to reduce the power consumption of Bulldozer which is higher than a 3960x at stock speeds. When you remember that over 1/3 of the transistors on the CPU (using the new 1.2 Billion transistor count) are in the L3 cache that only runs at 2.2 Ghz while the L3 on the 3960x runs at full-speed, you have to wonder at whether GloFo's 32 nm process has some fundamental flaws, or if AMD didn't listen to GloFo's design rules (or some of both).
2. AMD's and GloFo's combined marketing of their "gate-first" 32 nm process bragged loudly and repeatedly that gate-first (as opposed to gate-last used by Intel) gave 20%+ transistor density benefits and that Intel's process wasn't truly 32 nm. Well, when Bulldozer was reported to have a die area of 315 mm^2 and a 2 billion transistor count, this seemed like a justified advantage. Now, however, the transistor density of Bulldozer is lower than any other 32nm design from either AMD or Intel. Note: the same AMD PR guys that adjusted the transistor count confirmed that the 315 mm^2 die size is still accurate.
Rory Read is smart to shift the focus away from these unmanufacturable monsters and to put it on the next-generation of Bobcat and Trinity designs where AMD can actually leverage it's only real advantage over Intel: the GPU.
AntiFA: An abbreviation for Anti First Amendment.
They are not some bungled mess of transistors, they are Real Transistors(tm) this time.
Who's gonna start counting? Methinks its a PR excercise for all the shit amd are copping not being able to best a sandy bridge quad core with 1.5x as many transistors and (according to AMD measurements) 2x as many cores.
I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
If you bought a V10 car and it turned out to have a 4 cylinder, you'd be upset. No?
AMD just clarified that Bulldozer does have 2 billion transistors after all, but only 1.2 billion work. Which explains something about its performance.
I recall seeing that the top Bulldozer only had 8MB L3 cache, which seemed a bit low - Intel's equivalent top-of-the-line desktop models reach 15MB, and the server models 30MB.
At first, I just figured they were targeting the middle price bracket, but then they priced against the high-end. So I would not be surprised if much of the missing (or disabled, if that rumor turns out to be true) transistors belong to the cache.
I just paid 110 $ for my FX-4100. I want a refund ! Oh who gives a hoot is goes very fast( 4.5GHz) and I can't count that high anyway.
Those green panted stripes are actually paint over copper wires. Transistors are much smaller now than they were in the days of "Transistor radios" on top of that there are several types of transistors. The thing people need to notice here is that the processor performs the same task with less transistors than expected. Every transistor you remove from a microprocessor removes a small amount of leakage current. When you cut the amount of transistors nearly in half your are cutting the leakage current nearly in half. This is a good thing in many cases, it means that the processor should run just as well using less current, maybe even better. The less distance the signal has to travels (referring to how many transistors it has to pass through) the faster it will get through. This should mean you will see some speed up because of this. On top of that each transistor does take up a 'small' amount of room (measured in nanometers the measurement they measure molecules in.) on the chip, but 800mil still can take up a nice chunk of a processor. The nice thing about this is that it means they can cram more cores, graphics, or other useless junk onto your processor.
AMD PR is about 2 meter away, in the vertical dimension, from AMD Engineering.
Yet, AMD PR is asking "Reviewers" to tell them, AMD PR, how many transistors the AMD Bulldozer has!
Why cannot AMD PR send someone to AMD Engineering and ask AMD Engineering how many transistors is in AMD Bulldozer? Is such a National Secret? Is AMD PR afraid? Is AMD Engineering under court order not to talk to AMD PR? How much money does it take for a reporter at the Wall Street Journal to bribe someone at AMD PR to walk to AMD Engineering and ask this National Security Super Duper question?
Do you realize how many otherwise learned comments are tossing around the figure of 2 BILLION transistors on this CPU? Huh? Did anyone stop to wonder how many square mm 2 BILLION transistors would take up? I figure it would at least cover an entire ATX motherboard at extremely high density, and probably dissipate at least 50 kW of power.
It's MILLION, folks.
The 16core server parts were listed as having 2.4billion transistors at launch so either the FX PR was wrong/confused and it really is 1.2Billion transistors as it they say it is now for an 8Core FX or AMD manged to bolts on an entire second 8core processor to the server parts with 400million more transistors.
Anybody that has ever looked at the schematic for a VLSI chip at the schematic level will have problems figuring out what the transistors are for because so many of them are actually being used as resistors, diodes, or capacitors. Many are bias regulators or interstage coupling voltage level translators. Transistors are the simplest things to put on an IC so there tends to be lots of them. The transistor count rarely translates into a true level of complexity for the device over all. Having said that the last time a transistor count on a microprocessor meant anything was with Motorola's first two major processors. The MC6800 actually had about 6,800 transistors. The MC68000 had about ... wait for it .... 68,000 of them!
Good stuff: =ceiling(A1)
Bad stuff: =floor(B1)
...the future crusty old bastards are already drinking the Kool-Aid.