Opnion: Did We Expect Too Much of AMD Bulldozer?

Page 7 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Toggle sidebar Toggle sidebar
Status
Not open for further replies.
Yes, you expected way too much. As said before, this is an arch for the future. I wish that this debacle won't be the whole reason as to why the rest of you won't buy AMD in the future.

The truth is, there are multiple parties to blame. First is Global Foundries. They've been having trouble with the 32 nm SOI process for some time now. 32 nm SOI is not as easy as 32 nm Bulk CMOS (the process Intel uses). See Llano. Generating good enough yields is tough for them; AMD has been shipping Athlon IIs for FM1 sockets because the 32nm GPU half of the die is botched up and/or possibly dead.

Next is the media. Never in the "leaked" AMD slides did I ever see something about a guaranteed improvement over the competition. AMD never said they had the most powerful chip, they had the fastest. I immediately read that in terms of GHz when the articles started streaming out. To me, power is the amount of things it can tackle at the same time before breaking a sweat and speed is, well, speed.

Another one to blame is Intel. AMD just doesn't have the resources, flexibility, nor the market share that Intel has, not to mention the immediate dumb user bias has. I know of several people who I've built Intel builds for them who thought that it was all they needed (no GPU at all). I could have shoved in a C2 Conroe in there and they would have still thought that that was all they needed for playing games. Intel can botch up and still gain a good profit. AMD can't.

Finally and immensely, you the consumers. AMD needs money to fund their R&D. I predict that their only major cash flow is GPUs and APUs. Sure that makes them money. Sure they're pretty awesome at it. But what AMD needs is money to support the upper end of the PC spectrum. They just can't do it. If you want them to continue the development for high end CPUs, then they need to see the need for high end AMD CPUs. They know that in severs they to excellent but in workstations and home desktops it's really questionable.

I'll be skipping BD not for it's current mediocre benchmarks but because of financial issues. My Phenom II B55 works perfectly fine now. By the time PD is out, they'd have fixed the kinks out of the process and optimized the software. I just hope that the idiots who drone out "durrr it sawkz buing 2500K nao" would realize what AMD is looking forward to.
 
look how amd changed

http://www.tomshardware.com/review [...] 547-3.html

Patrick: The 6000+ is the fastest Athlon 64 X2 dual core processor ever, but what happened to the FX family?

Damon: Patrick, you are right. The X2 6000+ is the fastest AMD64 dual-core processor ever... so why isn't it called FX? To answer that I have to explain what FX is all about... pushing the boundaries of desktop PCs. FX-51 did that right out of the gate, with multiple advantages over other AMD processors, and a clear lead on the competition. Move forward a bit to where AMD put high-performance, native dual-core computing into a single socket with the FX-60. Fast forward again and you see FX pushing new boundaries as "4x4" delivers four high-performance cores with a direct-connect, SLI platform that is ready to be upgraded to 8 cores later this year.

So, with the FX brand now crowning a 4-core (and future 8-core) platform, it seemed logical that new dual-core solutions should reside under the X2 moniker.
 
[citation][nom]molo9000[/nom]It's actually a pretty neat and innovative CPU.The problem is software. Writing multithreaded code is not trivial and programmers are used to ever increasing single thread performance, so why bother writing efficient multi threaded code unless u actually are writing some critical productivity app?The FX8150 actually beats the i7-2600k in some benchmarks. Shame that most games can't use 8 cores.[/citation]

It's a MONSTER CPU. 2 Billion transistors. And they can only sell it for $245 and not competitive at $245 at that! This is going to destroy their bottom line in the CPU department. Especially since 2500K is nearly half the number of transistors and can do the same amount of work.

Sell AMD, but Intel. Nuff said.
 
I did not expect the BD to outperform the i7-2600K, i did expect it to be priced more competitively. I waited 4 months to buy my upgrade becuase of it's impending release. On Oct 12th i ordered my i7-2600K after reading reviews on the 8150. The core i7 cost me less than the 8150 would have.
 
I think we weren't trying to expect too much out of bulldozer - HOWEVER i would have at least expected 15% jump in performance AND/OR power efficiency when compared to any other AMD product - neither of which realized in this chip and that was the let down,,

If i could have bought a 8 core chip and easily beat the 1100 in all benchmarks (which it doesn't) or at least have a lower power usage (which it doesn't) I would have been estatic,, however this thing sucks back the juice and doesn't put out the numbers that make the difference for current AMD customers to "upgrade"

at least Intel gives you a compelling reason to buy 'all new' every couple years with each new design as each one makes the old ones look bad
 
I respect AMD and i can understand there stand point they are designing a future proof model and we have only yet to see the true performance enhancements as software starts using multiple threads, it is a matter of time before any of that happens. So i stick to them however i think they mde a bad judgment call for CPU industry of todays standard as the human race is we want it NOW not THEN! So that doesn't work in there favor however. I think its a good series of chip i just think the software architecture needs to catch a wake up and jump on the ship with multi core processing as they are taking there sweet time. AMD You Still Rock, also there arent enough benchmark tests utilizing all cores for example i do a lot of video rendering i know for a fact that an AMD octocore will outperform any Intel Quad Core for that interface. I do believe this processor series is for a particular target market and not for everyone especially gamers and people who utilize more single threaded apps then multithreaded apps.
 
More like BooDozer I guess if you are into video editing the it's a good buy other than that the Ivy Bridge is been promised a 20% performance gain over Sandy Bridge I'll wait for that
 
No buddy is thinking that bulldozer is next step to more advanced APUs. Just think about it when they incorporate a strong GPU for every pare of cores, they get both FPU and Int calculation power to the top.
 
I certainly did feel that AMD's marketing was trying to make Bulldozer sound great, but I tried to come in with as "fresh" of a mind as I could once benchmarks were released, and in fact, held off on buying a new computer until then.

Ultimately though, for what I need (where my opinion matters), AMD fell short of curve.
 
It would seem to me that this is a tactic that AMD has always done. They came out with 64bit extensions YEARS before Intel made them for desktop parts. Then they did dual core years before Intel as well. The same can be said about moving many traditional north bridge roles to the CPU die. All in all AMD is very progressive with their archatecture, bringing traditionally 'high end' or server grade features down to the home user.
Bulldozer follows in this line. When you look at being able to do super high end workloads on it, the thing just humms allong. The problem is when you try to do simple things, like browsing the web and playing games, it will be slower and more power hungry than everything within $75 of it. This is why Intel works with software companies like MS and Adobe to make sure that they know the direction that they are moving in order to make their products work well. AMD seemingly just comes out with a product, lays it on the world, and then expects the world to change the way they do things. It took until Vista to have a decent 64bit OS, and now it is going to take just as long for software developers to take advantage of Bulldozer. At least in the past AMD had provided kick-ass performance on normal workloads as well as the neat new features, but bulldozer is just bull.
 
Windows 7 is not ready for it, Windows 8 will address cores correctly. The fact is that, the new architecture is not what software developers could expect, so core management will improve it drastically. The real problem is, with just 4 FPUs, the FX-8000 series will never catch the Phenom II x6 performance many applications, and that would be predictable. Anyway, any kind of application is kinda OS dependant, so as long as the new Windows 8 corrects these flaws, we may expect better results for it (and this includes games in a minor way).
 
I think that for the moment, Zambezi is terrible but as single threaded apps die out that Bulldozer will gain in position in terms of benchmarks. I do wonder though how Zambezi handles multiple apps running simultaneously. That would be an interesting benchmark (at least to me) and to see a benchmark run in Linux. I have the feeling that Bulldozer will be sort of like the Pentium 4 in a way. It had potential in some segments of its architecture but, the problem was that when it first came out, developers didn't know how to use it! I do wonder how optimised versions of these tests would fare with Bulldozer's architecture though in the interest of seeing what is there because to be fair to BD, it *is* an entirely different architecture (more suited for the server market if my memory serves me correct).
 
[citation][nom]yankeeDDL[/nom]At $170 the FX-8150 would be a no-brainer. But that's 40% off the current price ...[/citation]
Hopefully the next generation won't be so dissapointing and we'll see some improvement. But again, for now what else can they do? this would be the only way, if the retail proccesor would be around 200$ WITH the stock liquid cooler included, i think it might be fair. For now 😀
 
YES, DEFINITELY, I think we expected far too much of BD. I'll "blame" the last four years of teasing and hype from AMD.

This reminds me of what Intel likely wants to totally forget, "Core" (yes, remember the Core line - not Core 2 line) which was crap compared to AMD CPUs of the time.

If Bulldozer had trounced Intel's current offerings without software that supports the specialized AMD instruction set, I am willing to bet the FX-8150 would be priced in the low $1,000 US range. Personally, I would not buy a CPU at that price.

In multi-threaded scenarios with the AMD extensions used and well-programmed, BD may be ahead of its time. For me, the performance bests Intel in areas that I care about. I am not a hard-core gamer, but I do run math intensive apps.

Though I won't buy an FX-8150 as I just updated one of my machines to a Phenom II 965 that is easily running at 3.8 GHz on air, when I replace my Opteron 1220 system next year, I will probably buy a BD related chip at that time.

I'll support AMD because they are the underdog. Without AMD in the market place, even Intel fanboys would complain at the high cost of their parts.
 
[citation][nom]dkraptor[/nom]Hopefully the next generation won't be so dissapointing and we'll see some improvement. But again, for now what else can they do? this would be the only way, if the retail proccesor would be around 200$ WITH the stock liquid cooler included, i think it might be fair. For now[/citation]
IMHO, that won't be long in coming. :)
 
Of course its a huge disappointment. Bulldozer benchmarks have been highly anticipated. While its not downright terrible, they are disappointing. Maybe in 6 months the prices will drop. At 149.00, it would be pretty sweet. At 245.00? No.

 
A+ article in my book.
where to start with all the points Wolfgang hit on:
1.) AMD marketing = FAIL
2.) true core/module debate = FAIL
3.) Chris Angelini = awesome review, much appreciated
4.) overclocking record on two cores with all others disabled = mis-leading and FAIL
5.) AMD handpicking results to advertise = FAIL
6.) AMD Athlon X2 7850 BE when overclocked @ 3.2GHz or better = a very very able chip,
I have one. WIN
7.) FX-8150 often gets beat out by the Phenom II X4 Deneb,
how can you release a new arch that gets out performed by the previous at all after 4 years.? = FAIL
8.) Pricing of the FX-8150 = FAIL

Did I expect too much.?
no.
I was saying all along that if it could at least beat out Intel / Nehalem then it would be a success to some sort.
well in my opinion, it doesn't.
so I did have better expectations yes, I will admit that but no.
I didn't expect to much and disappointed they didn't do better.
9.) AMD = FAIL

I will also say this, I'm not just going to throw out my present AMD units because of this fail.
the Phenom II X4 Deneb and other higher end AMD chips are still very suitable compared to this Bull-dud crap.
I do hope the FX-41xx series will be the best of the group and might be a somewhat upgrade
and more a side move from the Deneb.

Should of called it Phenom III..😉
 
I didn't expect to much, quoting myself from the "Amd breaks Speed Record":

[citation][nom]upgrade_1977[/nom]MHz doesnt matter, benchmarks matter. Nobody is gonna game everyday with LN2. Everyone knows MHz for MHz intel cpu's are more powereful, if they want to prove something do it at stock or equivalent clocks in benchmarks, or beat intel in price vs. Performance..[/citation]

I knew from the beginning this is just a marketing ploy. You don't have to be a genius to figure it out, it's all in the details, and fairly common knowledge that you have to compare benchmarks at the same clock speed to compare architectures, and intel has been a lap ahead damn near the whole race. AMD always goes for the marketing strategy vs. all out performance when it comes to their products. If you look at any of their product specs and don't know what your looking at, Ati and Amd products always look better on paper, but in the real performance world, normally they fall behind. Thats something I hate about AMD, and why I avoid them. Yes, i'm a fanboy, but I used to be an Amd Ati fan, just lost faith over the years. Now I'm all about intel and nvidia and haven't been happier.

I actually feel sorry for anyone who falls for Amd's marketing ploys, and runs out and actually buys this processor, just to find out it is just a mid range cpu. I mean, it gets beaten pretty badly by the i7 920 (AT STOCK SPEEDS!!) in some benchmarks, and that chip is over 3 years old. If I bought one based on there marketing, i'd be furious......FURIOUS!!! But thats just me. Thats why you should "Always" wait for benchmarks, and do your research before spending your money. Patience is a virtue.
 
This may be the straw that sends me to look at Intel for the first time in years for my next upgrade instead of sticking with AMD.
 
AMD should not be putting a server chip in a desktop right now. Of course, small, numerous cores would work well in a server situation, though.
 
Bailout said:
"Im surprised out of all these comments (only read two pages) no one is bringing up the fact that AMD blew their wad on purchasing ATI. Rather than investing in itself, they went ahead and foolishly bought ATI, and now they are paying the price in being a generation (or two) behind Intel CPU. "

ATI and AMD's bobcat are the only real success stories for them, I think ATI was one of the smarter moves AMD made.
 
Bailout said:
"Im surprised out of all these comments (only read two pages) no one is bringing up the fact that AMD blew their wad on purchasing ATI. Rather than investing in itself, they went ahead and foolishly bought ATI, and now they are paying the price in being a generation (or two) behind Intel CPU. "

ATI and AMD's bobcat are the only real success stories for them, I think ATI was one of the smarter moves AMD made.
 
to be honest, and i'm gonna get flamed for this, i'm laughing at all the "bulldozer is gonna blablabla..." comments right now. (I'm not a fanboy btw)
They fell flat on their collective asses, and that's it!
 
I thought AMD had said that bulldozer was going to compete with the i5 and not the flagship cpus? This was my expectation coming in so though I would have liked to have seen it blow Intel out of the water I am ok with the results.
SMH at some of you supposed AMD fanboys going out and buying Intel now, yeah this is why AMD can't beat Intel because Intel has a huge corner of the cpu market and has way more to spend on R&D. If you were true fans you'd buy AMD chips so that they would have more revenue to make better chips for the future.
 
Status
Not open for further replies.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom