Why is the response to Bulldozer so overwhelmingly negative?

[Gigantopithecus]

If the FX-8150 were priced the same as the i5-2500K, it would be a better value unless you were primarily a single-threaded application user. It's a harder sell at its current pricing. For gaming, it would be a moot point since anyone using these chips is gaming at 1080P and is GPU-bound, not CPU-bound. As the prices align now, the FX-8150 is not a particularly compelling value, unless you primarily use heavily-multithreaded applications.

Which is exactly why I already ordered two for my lab - it costs less than the i7-2600K and either rivals or bests it in modern multi-threaded titles. My genomics research uses modern software. But my uses are not typical of the average desktop user. Bulldozer clearly illustrates how a chip can absolutely dominate in one type of task (see Anand's 7-Zip benchmark) while downright stinking for another task (see any single-threaded application benchmark). Five years ago, everyone benefited from moving to a dual core from a single core. Today, not many people will benefit from moving from four to eight cores. My work computing absolutely will benefit from Bulldozer. My home computing absolutely will not. "What are you going to do with your computer?" is now more important than ever in determining what CPU will suit you best.

Are there really so many enthusiasts here and elsewhere who are so myopic they fail to see the flagship Bulldozer SKU beating the flagship Core SKU in a few relevant, real world applications? Are there really so many failing to see that while Bulldozer did not wrest the performance crown from Intel, it finally brings some semblance of competition to the i5-2500 and i7-2600? Again, some people are going to benefit from Bulldozer over the high-end Core SKUs. But not everyone.

Finally, where are the reviews of the $165 FX-6100 vs. the $180 i5-2300? ...What about the $115 FX-4100 vs. the $125 i3-2100? First, you can actually overclock the AMD chips. You can't overclock the i5-2300 nor the i3-2100. Second, Anand's review made it very clear that Turbo Core in the Bulldozer CPUs works really well - better than it did in the 6-core Thubans, and better than Intel's Turbo Boost. What effects does this have in typical real-world usage scenarios given the i5-2300's comparatively anemic and i3-2100's non-existent Turbo Boost? There are a lot more people who drop <$200 than >$200 on CPUs - and right now, do we have definitive knowledge of the specific, relevant comparisons I mentioned? Why would anyone dismiss an entire architecture when really all we know is how its high-end compares to the competition's high end?

 

[biostud]

maybe because 99&#37; of users are better of with SB while the last 1% are better off with BD

 

[Piano Man]

Hey I like unzipping files as much as the next guy, but I don't buy a CPU based on that.

 

[Patrick Wolf]

maybe because 99% of users are better of with SB while the last 1% are better off with BD

:thumbsup:

 

[Ken g6]

2500k (and 2600k) can overclock more, use less power, and put out less heat. Power and heat do matter - electricity does cost money.

Finally, where are the reviews of the $165 FX-6100 vs. the $180 i5-2300? ...What about the $115 FX-4100 vs. the $125 i3-2100? First, you can actually overclock the AMD chips. You can't overclock the i5-2300 nor the i3-2100. Second, Anand's review made it very clear that Turbo Core in the Bulldozer CPUs works really well - better than it did in the 6-core Thubans, and better than Intel's Turbo Boost. What effects does this have in typical real-world usage scenarios given the i5-2300's comparatively anemic and i3-2100's non-existent Turbo Boost? There are a lot more people who drop <$200 than >$200 on CPUs - and right now, do we have definitive knowledge of the specific, relevant comparisons I mentioned? Why would anyone dismiss an entire architecture when really all we know is how its high-end compares to the competition's high end?

These I'm more interested in. Of course, that's the low end again; everyone on enthusiast sites like this one was hoping AMD would compete at the high end.

 

[cotak13]

maybe because 99&#37; of users are better of with SB while the last 1% are better off with BD

I think it should be 99% of people think they are better off with the SB. When in reality for that 99% most of the time they aren't better off with either and should have saved their money buying a cheaper processor from either company.

People who push their computers hard are a minority in reality. Most people use their computers like how their grandparents uses a tv, as an entertainment/distraction tool.

The battle ground for consumer is moving to laptops anyhow. Fewer and fewer are buying desktops. The other battleground is in server. I have yet to see anyone do a server bench on the new chip. I'd be interested to see what the server BD is like for server work loads.

 

[Gigantopithecus]

maybe because 99% of users are better of with SB while the last 1% are better off with BD

Then that leads to the simple question of what are people doing with their $200 and $300 i5-2500K and i7-2600K CPUs? The FX-8150 either rivals or bests the i5-2500K in Cinebench. FX-8150 flat out stomps i5-2500K in 7-Zip, which I've found is a great proxy for many scientific analysis applications (like the ones I use). AMD beats Intel in TrueCrypt. FX-8150 closes the gap in PS. FX-8150 beats 2500K in Monte Carlo simulations, which are enormously important in a number of research fields from economics to evolution.

Again, I'm not making the argument that Bulldozer is better than Sandy Bridge - it clearly is not. But what are the enthusiasts here and elsewhere doing with their computers that makes so many people think Bulldozer is a failure? I would think we'd be more likely than anyone to recognize what's good about Bulldozer. ...Really, if you're not into HPC - why are you buying an i7-2600K anyway?

And where's Johan's server review?!

 

[Ancalagon44]

maybe because 99% of users are better of with SB while the last 1% are better off with BD

99% are better off with SB, 0.999% are better off with Phenom II (if they already own an AM3 board), and 0.0001% are better off with BD.

BD is owned at everything that matters to me by its predecessor, and thats even including power consumption.

 

[Snoop]

Finally, where are the reviews of the $165 FX-6100 vs. the $180 i5-2300? ...What about the $115 FX-4100 vs. the $125 i3-2100?

Seeing the speed of the 8150, I imagine the 4100 and 6100 are dog a## slow. I would be surprised if either of them could hang with a phenom II quad.

 

[Gigantopithecus]

I think it should be 99% of people think they are better off with the SB. When in reality for that 99% most of the time they aren't better off with either and should have saved their money buying a cheaper processor from either company.

THIS. People frequently ask me why if I'm so interested in PCs, then why do I use an i3-2100 at home? The answer is...

People who push their computers hard are a minority in reality. Most people use their computers like how their grandparents uses a tv, as an entertainment/distraction tool.

The battle ground for consumer is moving to laptops anyhow. Fewer and fewer are buying desktops. The other battleground is in server. I have yet to see anyone do a server bench on the new chip. I'd be interested to see what the server BD is like for server work loads.

Bingo. I knew I wasn't the only one wondering what's going on with the Bulldozer response, ha.

 

[Gigantopithecus]

BD is owned at everything that matters to me by its predecessor, and thats even including power consumption.

Are you 99% of computer users?

 

[Ancalagon44]

Are you 99% of computer users?

Uhh yeah. I play a few games, run Firefox and winamp, run OpenOffice now and then, and sometimes a Netbeans based IDE and compilation of small Java projects. Why on earth would I need a CPU built for scientific or financial calculations?

For all of the above, an Intel CPU would be faster. Heck, a 2500K is even cheaper than an 8150. It also draws less power, especially overclocked. Remind me why I would buy BD?

 

[Vesku]

I expected a 2 billion transistor budget chip to perform a bit better vs it's own companies 0.9 billion transistor chip (Thuban x6). Given that it was a next gen design why is single thread IPC less than a Phenom II? It makes sense that due to shared design it would be less IPC than a theoretical Phenom III, but Phenom II?

 

[Red Storm]

I too would like to see a review of the lower end BD chips compared to similarly priced Intel chips. Anyone with half a brain knew AMD wasn't gonna take the performance crown, so I'm interested to see how they faired in the low to mid level market.

 

[Gigantopithecus]

Uhh yeah. I play a few games, run Firefox and winamp, run OpenOffice now and then, and sometimes a Netbeans based IDE and compilation of small Java projects. Why on earth would I need a CPU built for scientific or financial calculations?

For all of the above, an Intel CPU would be faster. Heck, a 2500K is even cheaper than an 8150. It also draws less power, especially overclocked. Remind me why I would buy BD?

You shouldn't buy an FX-8150. You shouldn't buy an i5-2500K, either. Without knowing what games you play, based on what else you describe, you would be fine with a $60 AMD Athlon II X2 250.

Again, why would anyone think "Bulldozer is a failure because its capabilities are not what I would take advantage of?" That seems to be the overwhelming sentiment here and elsewhere. That's like saying semitrailers suck because all you do is run to grocery store a few times every week.

 

[Ancalagon44]

You shouldn't buy an FX-8150. You shouldn't buy an i5-2500K, either. Without knowing what games you play, based on what else you describe, you would be fine with a $60 AMD Athlon II X2 250.

Again, why would anyone think "Bulldozer is a failure because its capabilities are not what I would take advantage of?" That seems to be the overwhelming sentiment here and elsewhere. That's like saying semitrailers suck because all you do is run to grocery store a few times every week.

I run an Athlon X4 620. A 2500k would help load and boot times and also framerates depending on the game.

I dont get what you are saying. 99&#37; of users have loads similar to mine. BD is a complete failure at those tasks. It is great for you, but you arent the average user are you? I'm happy that BD is suitable for you, but for the other 99%, its a complete failure. Its such a bad failure that it would literally be a better idea for me to buy a Phenom II X6. That is how bad it is. Doesnt that scare you? It cannot beat its predecessor?

And anyway, what you are saying is that the Porsche you are driving around in would be great to deliver furniture in because you like driving it on the track. I'm glad its good on the track but it aint no good for furniture.

EDIT: If I had to emphasive one thing, it is that BD is a failure because the now ancient architecture of Phenom II outperforms it in every way. I dont mind if Phenom II outperforms Bobcat but it should not outperform BD. Same target market = user experience should improve. It does not.

 

[cotak13]

Then that leads to the simple question of what are people doing with their $200 and $300 i5-2500K and i7-2600K CPUs? The FX-8150 either rivals or bests the i5-2500K in Cinebench. FX-8150 flat out stomps i5-2500K in 7-Zip, which I've found is a great proxy for many scientific analysis applications (like the ones I use). AMD beats Intel in TrueCrypt. FX-8150 closes the gap in PS. FX-8150 beats 2500K in Monte Carlo simulations, which are enormously important in a number of research fields from economics to evolution.

Again, I'm not making the argument that Bulldozer is better than Sandy Bridge - it clearly is not. But what are the enthusiasts here and elsewhere doing with their computers that makes so many people think Bulldozer is a failure? I would think we'd be more likely than anyone to recognize what's good about Bulldozer. ...Really, if you're not into HPC - why are you buying an i7-2600K anyway?

And where's Johan's server review?!

I think you know the answer. For most of these people they just want the "fastest". It's like why do people buy cars with 300+ horsepower? Some would argue that in North America it's useful for the highway on ramps (in reality it's only useful once in a blue moon, and dangerous at other times). For most people they want it because they "feel" it's better.

And the sentiment for these tendencies are of course fed by their related media. As that's how they make their money, getting people pumped about new things and hitting their pages often.

Most people here knows and will tell you that the 2500k is a better choice over the 2600k for gamers because well it cost less and give you the same performance in most games. Yet when they get a BD review that doesn't beat the 2600k somehow it's a problem.

I think it's partly due to the expectations. People wanted a new revolution in performance. And when they don't get it they pout like a little kid and sulk about it for a while.

 

[996GT2]

Are there really so many enthusiasts here and elsewhere who are so myopic they fail to see the flagship Bulldozer SKU beating the flagship Core SKU in a few relevant, real world applications? Are there really so many failing to see that while Bulldozer did not wrest the performance crown from Intel, it finally brings some semblance of competition to the i5-2500 and i7-2600? Again, some people are going to benefit from Bulldozer over the high-end Core SKUs. But not everyone.

I don't think anyone is being myopic when they consider Bulldozer a failure. Here's my opinion:

There's the issue of power consumption, which quite frankly is atrocious on an FX-8150.

There's also the issue that the FX-8150 costs $280 at Newegg currently. Well above the pricing of the 2500K which beats it in just about every benchmark while consuming less power. If the FX-8150 was $200, I could see it being a moderately good deal. At $280, and with the poor performance + horrendous power consumption, what's there to like?

 

[Hypertag]

You shouldn't buy an FX-8150. You shouldn't buy an i5-2500K, either. Without knowing what games you play, based on what else you describe, you would be fine with a $60 AMD Athlon II X2 250.

Again, why would anyone think "Bulldozer is a failure because its capabilities are not what I would take advantage of?" That seems to be the overwhelming sentiment here and elsewhere. That's like saying semitrailers suck because all you do is run to grocery store a few times every week.

CPUs don't matter for games is the spin you want to go with? http://www.xbitlabs.com/articles/cpu/display/amd-fx-8150_10.html#sect0

"Bulldozer is a failure because its capabilities are not what I would take advantage of?""

What capabilities? It uses eight cores to barely beat the quad core 2600k in 7zip. If we are suppose to be cheering in the streets because 2 billion transistors barely outperformed 900~ million transistors while consuming twice the power, then I don't get it . Hell, compare it to the 1 billion transistor 1100t. The extra billion transistors bulldozer has over the 1100t are used to provide performance that is between 15% worse and 15% better than the 1100t.

 

[Patrick Wolf]

Again, why would anyone think "Bulldozer is a failure because its capabilities are not what I would take advantage of?"

Because those capabilities are tiny and a comparable Intel CPU is good enough and much faster for everything else. Don't think most people using a desktop are using it exclusively for one or two specific tasks.

 

[Ferzerp]

So even though we've had to put up with bulldozer thread after bulldozer thread, it isn't going to stop now that it's obviously a massive flop. (Well, it was obvious months ago really.)

Now we are going to get endless fanboy damage control threads? Here is a tip, self-identify with a faceless corporation less, and you will actually be able to make smart buying decisions.

 

[Vesku]

Gigantopithecus, when you get your 8150s up and running please post a comparison for your purposes versus a 1100T.

Will be interesting how much more performance you get out of more than twice the transistors and the corresponding higher power draw needed to make them work.

 

[996GT2]

Gigantopithecus, when you get your 8150s up and running please post a comparison for your purposes versus a 1100T.

Will be interesting how much more performance you get out of more than twice the transistors and the corresponding higher power draw needed to make them work.

Exactly. In terms of Per-core IPC, Bulldozer is actually SLOWER than Phenom II.

That above fact is downright sad in itself, but when you consider that Bulldozer also has much higher power draw than Phenom II and Sandy Bridge, it's even more sad. I (and most others) didn't think it was possible for AMD to fail this hard, but they did, and here we are:

 

[Gigantopithecus]

I don't think anyone is being myopic when they consider Bulldozer a failure. Here's my opinion:

There's the issue of power consumption, which quite frankly is atrocious on an FX-8150.

I think that's a very good point, and obviously, Bulldozer looks terrible compared to Sandy Bridge. How much of that is due to the superiority of Intel's fabrication technology compared to GloFo's? This isn't a leading question, I honestly don't know.

CPUs don't matter for games is the spin you want to go with? http://www.xbitlabs.com/articles/cpu/display/amd-fx-8150_10.html#sect0

No, that's not what I said. I have no idea what games they're playing so I wasn't about to comment on an unknown.

"Bulldozer is a failure because its capabilities are not what I would take advantage of?""

What capabilities? It uses eight cores to barely beat the quad core 2600k in 7zip. If we are suppose to be cheering in the streets because 2 billion transistors barely outperformed 900~ million transistors while consuming twice the power, then I don't get it . Hell, compare it to the 1 billion transistor 1100t. The extra billion transistors bulldozer has over the 1100t are used to provide performance that is between 15% worse and 15% better than the 1100t.

I'm torn between agreeing with you - that is an excellent interpretation - and not caring, because for now, my research will take less time to finish, power draw be damned.

 

[GoStumpy]

I suppose I'm just a positive thinker, but perhaps there isn't much (if at all) optimization for a brand-new CPU like the BD?

Sure SB has ownership of all current "benchmarks", but what about a few years from now? Software has to keep up with hardware, we all know that right now hardware is advancing faster than the software that utilizes it...

My other train of thought is:

Wouldn't 2 billion transistors be good for SOMETHING? I can't see AMD putting out a 2 billion transistor chip without seeing a massive improvement.

Seeing the overwhelming lead in 7-Zip makes me wonder if in time it will do the same performance in other applications?