AMD FX-8150 Bulldozer finally tested

[sin0822]

They might even keep unclocked chips in the mainstream segment, but they probably wont push IB stock clocks much over what we have today if they dont need too (3.6-3.8GHz 4C/8T with 5-10% IPC boost).

ivy bridge will have a 63x max Intel multi instead of 57x so i am guessing w will see like 65x

[SKYMTL]

The reactions of some people really get me laughing.

People wanted to see results posted so Lab501 did that. Unfortunately, some of us are bound by NDA so we can't comment on the accuracy of said numbers but....

Since they weren't anywhere near what some people expected, small details are picked and prodded until we get the "impossible", "BS!" and "biased" words thrown around for heaven knows what reason.

So if these numbers are accurate, will those same people post again saying "we were wrong!"? Methinks not. They'll go sulk in a corner for 5 minutes, find another thread to pick apart and then ultimately go buy an Intel processor. On the flip side of that coin, if the numbers are inaccurate or wrong, Lab501 has pretty much burned every possible bridge with a major manufacturer and will be the butt end of communal jokes until Theo or Charlie screw up again.

So which one will it be? Man, the next few days will be interesting to say the least.....

[Sparky]

So if these numbers are accurate, will those same people post again saying "we were wrong!"? Methinks not. They'll go sulk in a corner for 5 minutes, find another thread to pick apart and then ultimately go buy an Intel processor.

I disagree. I'll go sulk in a corner for 5 minutes, pick apart another thread, then continue using my current CPU

[PerryR]

Since they weren't anywhere near what some people expected, small details are picked and prodded until we get the "impossible", "BS!" and "biased" words thrown around for heaven knows what reason.

This line makes me think you know it's bad, since the same could be asked of people who know that'll it suck based on this one review

Regardless, I still find it strange that AMD is willing to charge around $280 for a failed processor. The price came from this microcenter employee.

http://www.overclock.net/amd-general...ate-specs.html

[xVeinx]

AMD stated nothing over $250 during the live event, so I imagine that pricing will change (eventually).

[SKYMTL]

This line makes me think you know it's bad, since the same could be asked of people who know that'll it suck based on this one review

I will repeat what others said: never, ever base your purchasing decisions or opinion about a certain product off of:

A) Rumors

B) A single review / preview


At this time everyone wants to know how this thing does so let's all wait until we see a large cross section of results before flinging crap around.

[qurious63ss]

IIRC the PII 1090 and 1100 where both around 290 when first released but dropped down to 190 pretty fast so you will probably see this for these proc price drop as well. Supply and demand will dictate the price. Now if current results hold true I guess I should be glad Amd screwed up but I feel a bit a little bad for them since I know how much hard work it takes to get these things out successfully. Well if it helps I can talk to the big shots and see if we can fab those bulldozers on a proper 32nm process

[PerryR]

I will repeat what others said: never, ever base your purchasing decisions or opinion about a certain product off of:

A) Rumors

B) A single review / preview


At this time everyone wants to know how this thing does so let's all wait until we see a large cross section of results before flinging crap around.

I was just joshing tis all.

AMD stated nothing over $250 during the live event, so I imagine that pricing will change (eventually).

I've heard about this event, but can't tell which one it is.

[Gener_AL (UK)]

I have been lurking, waiting, and I have to say.
It's the strangest build up to a CPU launch I have experienced so far.
Although I am full of a fever at the moment so everything is a little odd.
Can anyone remember AMD being so tight lipped on figures prior to launch?
Just about anything bulldozer related.
I have spent the day testing various BIOS on AMD chips > compatible code > optimised > non-optimised.
The results that have leaked so far give indication that the real performance is yet to be unleashed, the previews I'm seeing here sort of confirm this data pattern.
I think first of all BIOS is the key here, followed by OS updates and then software optimisation.
if this is raw performance on a compatible but non optimised BIOS I am quite happy then.

[Kobaltrock]

If all these benchmarks are legit (and, at this point there is very little reason to doubt them since multiple people got pretty much the same results), the the only thing we are going to see the next few days is lots of spin control.

JF-AMD will have his work cut out for him, going into full damage control mode.

Nobody wants to only have one player in the mid to high end field, and at this time, it looks like that intel doesn't have anything to worry about, and they will continue to charge a premium for what they offer.

I still think AMD will do OK with the BD, they just need to adjust the price down, and hurry up and tape out pile driver.

[M.Beier]

Here is what I saw in actual gameplay with the settings that I actually use not some ty canned bench like what Hornet just posted like thats supposed mean something to me.

This is what I would expect from a "professional review" I should have left out the maximum framerate though.

This is an i5 750 at 4ghz and an i5 2500k at 4.7ghz.

And here is what I would have expected from you....
Preview, review, do you know the difference? I think you do, do you need glasses then?

[M.Beier]

The reactions of some people really get me laughing.

People wanted to see results posted so Lab501 did that. Unfortunately, some of us are bound by NDA so we can't comment on the accuracy of said numbers but....

Since they weren't anywhere near what some people expected, small details are picked and prodded until we get the "impossible", "BS!" and "biased" words thrown around for heaven knows what reason.

So if these numbers are accurate, will those same people post again saying "we were wrong!"? Methinks not. They'll go sulk in a corner for 5 minutes, find another thread to pick apart and then ultimately go buy an Intel processor. On the flip side of that coin, if the numbers are inaccurate or wrong, Lab501 has pretty much burned every possible bridge with a major manufacturer and will be the butt end of communal jokes until Theo or Charlie screw up again.

So which one will it be? Man, the next few days will be interesting to say the least.....

True, double true.
We want results, but fanboyism gets in the way for some..... Making Lab501 the bad guys for serving the enthuciasts..

[Ice009]

Monstru, I hope you not playing with us and the benchmark are true or not paid by Intel to do this. If its fake you desserve a big punch in your face from everyone who uses a AMD in this forum. Im not the happiest man in the world right now

Maybe he's been paid by AMD to leak it so if the actual performance is very close to the 2600K people will be ecstatic, rather than mildly disappointed if they hadn't seen these benchmarks .

[Viking1962]

Monstru, I hope you not playing with us and the benchmark are true or not paid by Intel to do this. If its fake you desserve a big punch in your face from everyone who uses a AMD in this forum. Im not the happiest man in the world right now

[god_43]

...lol wtf how did your comment end up quoted before you made it? who here is versed in temporal mechanics?

[mAJORD]

all the conspiracy theories are messing up spacetime

[Ice009]

I don't know how how my post ended up above his, the quote was done before the edit too .

[thebluemeanie1]

Lab501.ro +1

At least they left plenty of numbers for them to upgrade to.

The rest of you guys just remind me why I stopped visiting this site...

[Russian]

The reactions of some people really get me laughing.

People wanted to see results posted so Lab501 did that. Unfortunately, some of us are bound by NDA so we can't comment on the accuracy of said numbers but....

Since they weren't anywhere near what some people expected, small details are picked and prodded until we get the "impossible", "BS!" and "biased" words thrown around for heaven knows what reason.

So if these numbers are accurate, will those same people post again saying "we were wrong!"? Methinks not. They'll go sulk in a corner for 5 minutes, find another thread to pick apart and then ultimately go buy an Intel processor. On the flip side of that coin, if the numbers are inaccurate or wrong, Lab501 has pretty much burned every possible bridge with a major manufacturer and will be the butt end of communal jokes until Theo or Charlie screw up again.

So which one will it be? Man, the next few days will be interesting to say the least.....

I've been tempted to comment on some of this that's been going on but I'm in the same boat as you. Oh well.

[zalbard]

I've been tempted to comment on some of this that's been going on but I'm in the same boat as you. Oh well.

You could give us a hint!

[kuroikenshi]

[mongoled]

You could give us a hint!

Yes, please do,

would like to know if their is any light at the end of the 5 year tunnel!

[looncraz]

Part of me still hopes it might be a Radeon 9700 pro deal, where it was released and was sucky due to some serious stability issues (my brother had one at the time and experienced this), then they realized that the card hated AGP fast writes for some reason (which they did fix with the 9800). Disable fast writes in the system BIOS and the card turned into a monster that completely caught nvidia off guard.

Not likely, but I'm still sort of hoping there is some flaw in the BIOS coding or whatever that is easily fixed, turning the CPU into a monster.

Vain hope? Probably.

Actually, we already have such an issue known for Bulldozer, and NO bench-marked system has the patch installed!

The shared L1 cache is causing cross invalidations across threads so that the prefetch data is incorrect in too many cases and data must be fetched again. The fix is a "simple" memory alignment and (possible)tagging system in the kernel of Windows/Linux.

I reviewed the code for the Linux patch and was astonished by just how little I know of the Linux kernel... lol! In any event, it could easily cost 10% in terms of single threaded performance, possibly more than double that in multi-threaded loads on the same module due to the increased contention and randomness of accesses.

Not sure if ordained reviewers have been given access to the MS patch, but I'd imagine (and hope) so! Last I saw, the Linux kernel patch was still being worked on by AMD (publicly) and Linus was showing some distaste for the method used to address the issue. One comment questioned the performance cost but had received no replies... but you don't go re-working kernel memory mapping for anything less than 5-10%... just not worth it!

[haylui]

You could give us a hint!

don't!
later the hint goes against the fact...

[Hondacity]

Actually, we already have such an issue known for Bulldozer, and NO bench-marked system has the patch installed!

The shared L1 cache is causing cross invalidations across threads so that the prefetch data is incorrect in too many cases and data must be fetched again. The fix is a "simple" memory alignment and (possible)tagging system in the kernel of Windows/Linux.

I reviewed the code for the Linux patch and was astonished by just how little I know of the Linux kernel... lol! In any event, it could easily cost 10% in terms of single threaded performance, possibly more than double that in multi-threaded loads on the same module due to the increased contention and randomness of accesses.

Not sure if ordained reviewers have been given access to the MS patch, but I'd imagine (and hope) so! Last I saw, the Linux kernel patch was still being worked on by AMD (publicly) and Linus was showing some distaste for the method used to address the issue. One comment questioned the performance cost but had received no replies... but you don't go re-working kernel memory mapping for anything less than 5-10%... just not worth it!

wow