I don't know how would you define AMD having an inefficient cores compared to Intel. There never was a comparison between inefficiency vs efficiency. A single BD core does not take as much of a die space as a single SB core. If you were to compare die space which is a rough measurement of transistor count, a single SB core will be as big as a single BD module (2 cores) minus L3 cache. So therefore I can safely say that if a process can only use single thread, SB is a clear winner here but if it could use two, BD has an advantage.

Of course some might claim that it is just splitting transistors to make it two cores but if it is never that simple to do that. There are downsides of having more cores definitely but I'd rather get a SB that has 4 cores instead of using the same transistor budget for one huge SB core. There are diminishing returns if you want to make a processor more complex. Designing a smaller core and copy paste it 4 times is also much more economical than designing a single huge monolithic core.

What AMD wants with BD is that it is like Intel's HT but with HT you're looking like a <20% gain in a best case scenario. There is no replacement for physical cores and AMD's approach is to have two cores that perform 80% as good due and have them work in parallel. So in a two threaded workload, a module should be better than a single SB core with HT. I won't say that AMD's approach is good or bad but there will be a market where parallel workloads are important, mostly involving servers. The statement below is not written by me but it should give you a clear idea of what I mean.
Not Fusion optimized, Avivo is dealing with videos and what I mean by OpenCL or what AMD calls it AMD APP, you can use the GPU for just about any parallel workloads and it does not have to be video based only.

They have fallen behind quite a bit but we cannot compare what most consumers are willing to pay nowadays since most are quite thrifty after the recession.

I don't have much info about compatibility of IB with the current Panther Point but most do suggest that there is compatibility but I'll wait till Intel gives the green light.

C language is not optimized for multithreaded so a rewrite to OpenCL is definitely necessary. They are few now but I'm sure they will be more popular as it gains momentum. I'll see how C++0x deals with parallelism.

I wish I have extra time to read through Nvidia's 61 pages and AMD's 142 pages of the OpenCL programming guide.

Give it some time. AMD is a small player and when Intel adopts OpenCL, we'll see a lot of OpenCL programs. In fact Intel has their OpenCL SDK already.

I agree that not all can be GPGPU accelerated but some workloads can be done by the GPU. Scaling for CPU performance with # of cores is pretty bad especially if the code written for it is not optimized for multicore.

Nvidia is the current leader for such standards no doubt about that. AMD is just starting with their HD7000 series or their current iteration of their GPU architecture is more compute based.



This post has been edited by dma0991: Jul 28 2011, 05:01 PM

I won't know for certain how well Intel MIC will perform in real world yet but from my understanding, if a portion of a code is not parallel enough, it will be subjected to Amdahl's Law. Without the code being as parallel as the hardware, it will reach to a maximum number of cores and there is no performance gain afterwards. I am not sure how parallel is Intel's implemention with MIC but I think if the programming language from the ground up is made for parallel, it should be able to benefit the most. This is my opinion only though and might not reflect how well MIC will perform.

AVX lol, another 500+800 pages from Intel that I have not read. AVX is also implemented with BD. I do not know much about Larabee but from what I have heard others have said that it is not really awesome. Like I said, market dominance can change with time and since AMD is going towards that direction, Nvidia is going to have a little bit more competition than they used to have. Intel MIC, maybe.

September 19th. Jot that down on your calendar or you could pre-order now and be the first to get when it comes out.

AMD Bulldozer Enhanced (FX-Komodo) Processors Will work on AM3+

I think that should be the way if FM2 has the same pin count and pin position as FM1. There should be backwards compatibility if it is true and I do think AMD wouldn't be changing sockets without backwards compatibility with every processor generation they produce.

1. Not very likely that SB will have a price drop although it is speculated to happen. Intel could easily just sell a refresh with higher stock clock speeds or BD is not going to be for every user so it doesn't bother Intel at all that they should reduce their prices.
2. Stick with AM3+ boards that have the 970 or 990 chipsets and a full sized board as I have seen some of the CPU support list for the Gigabyte's range and lower end boards will only support 95W TDP Bulldozers while higher tier boards will be able to support all. Could be applicable to other brands as well.

1. Supposedly there are rumors saying that there will be a SB refresh to counter BD at launch but it is still a rumor only. SB is the next best thing to BD I think as there are rumors of Ivy Bridge being delayed as well. Pricing depends but should be around RM900 +-
2. Not necessarily high end. Some boards cant support a 125W TDP processor . Just double check the CPU support list of the board you're buying if you intend to get the top end FX-8150.

I have no idea but they should release it soon. Rumored to be 19th September but starting to sound less and less likely. We'll see what comes by then.

AMD Reschedules Launch of FX-Series Chips to October, Changes Launch Lineup

It is in mid October or Q4 2011 this time compared to 19th September. FX-4170 also does not look promising with such a chip having a 125W TDP @ 4GHz+ as it might indicate there is not much overclocking headroom compared to the Core i5 2500K.

The Start of a New Era

It is confirmed that there will be nothing this 19th September. Q4 is October onwards.

Sadly BD might not be the holy grail for the ultimate gaming rig that everyone is looking for. BD is more optimized for server like workloads and in a server where VM and heavily threaded application scales very well with core count, the regular consumer's PC is not so. Single threaded performance for BD is out of the question and higher clock speeds doesn't mean much when it's IPC is expected to be much lower than SB and much closer to Nehalem.

Most likely AMD is abandoning the enthusiast consumer as the market seems to be shrinking lately due to the majority of users want a handheld device. They have Llano to attack the mobile space and BD to attack the server space that will be providing service like cloud to mobile devices. The direction is still uncertain as AMD was under lackluster management by Hector Ruiz and Dirk Meyer. Hopefully under Rory Reed, things may change for the better in terms of competitiveness.

I can't say for sure that the enthusiast market is shrinking but from what market analysts say, the forecasted growth in percentage has dropped therefore there is a sign that it is shrinking. As much as I hate to say this, the enthusiast market is actually quite small when you compare with the number of mainstream consumers. There are more people who are illiterate when it comes with computers by a huge margin than people who are very technical with their PC.

Some say that SB is the sweet spot for a enthusiast to be and IB might not promise much either as it is geared for the mobile space. All the 3D transistor you heard of has more benefits towards the mobile space and Intel is not sitting down when it comes to mobile space either when they can dump $300 million to make sure that Ultrabook is a success. Probably desktops will gain some popularity once again when the handheld device bubble pops. There are only so many cores you can put into a handheld device and scaling is an issue when there will be a quad core ARM processor in the making this year, all of this happening while battery capacity has not improved at all.

You do realize that voice recognition is not exactly a strong point for a computer. Computer language is rigid while human language is not. We can mix and match what we say and it will still be understandable to the other person but not so for the computer. To actually push for voice recognition you would need a database of voices and of which Google has done some of the work but it is definitely their patent and not willing to sell them. Even so, I don't think you can do speech recognition with the phone alone only.

I tried doing the speech recognition on a Galaxy S before and I don't think it works while being offline which leads me to think that your voice is relayed to Google's server before relaying back as a translation. Even speech recognition is rigid in a sense that some needs a starting word before a command is given. Tablet and such are entertainment devices and hardly any serious work can be done on them. A real keyboard is there to stay for a very long time. I'd like to see someone actually try to type without looking at their touch screen, I'm sure there are who could but the absence of tactile feedback makes that much harder.

No, I'm not saying that speech recognition is not a strong point for a computer, it definitely will give an added advantage to a computer. I'm saying that based on how a computer works, it can't really recognize speech directly like how humans could. It is not so much that speech recognition has not been progressively researched, it actually is a work in progress. I had friends attending an even that is hosted by Google for speech. They were given a Google Nexus, a piece of paper with certain words to say while the Google Nexus records their voice. It is part of how speech recognition works and you would definitely need a database on speech types as there are infinite variation to voices. The sad part is that I did not get to join the event, my friend was given a nice Google t-shirt which I want.

We do have enough hardware to process but not with the smartphone or handheld itself. We still have to rely on a large database and large amount of processing power of a server to be able to translate fast enough. The algorithm for speech recognition is relational in one way or another and something like IBM Watson which is almost similar to speech recognition. It can't translate what it doesn't have in it's database. You would need a large enough database of required data and fast enough processing power to make sure that an answer is given instantly.

Nice one.

It was made from the ground up to have high clockspeeds as AMD knows that they don't have the IPC to match SB. Nevertheless I heard as well that overvolting till 1.5V is still quite a safe range for BD.

Regardless of how BD will actually perform it actually better for AMD if they do not show any results prior to launch. They are small compared to Intel and showing your trump card early in the game is going to hurt AMD's chance of making sure that their competition doesn't know how well their products perform. No matter how well BD will perform, there is no way that AMD will cause Intel to lose their share of the market period. No matter how good BD will be, you wouldn't see AMD being the market leader and Intel as the underdog in a short period of time. Also, GloFlo does not have the production capacity or R&D budget to actually be as competitive as Intel.

You shouldn't reveal a product before launch, read about Osborne Effect.
3.8GHz with your stock voltage, not the average stock voltage that all processors have which varies depending on whether you're getting a lemon or a golden chip. If Intel were to release SB at a higher clock speed, the yield will drop as the bar for standard quality is raised, it has to be at the right range based on their internal batch testing.

You can't compare Intel and AMD directly with IPC as both are using different methods to achieve the same goal. SB has higher IPC but more likely lower clock speed, BD has higher clock speeds to make up for the lower IPC. That is why BD can break the speed record because it is like Netburst, which Intel is no longer pursuing the high clock speed and low IPC type.


Anything OBR says, write or does have to be taken with a large chunk of rock salt, a grain wouldn't be enough.

Not too sure about your SB but on average, most wouldn't go over 4.5GHz for a stable 24/7 OC which is quite near to the 1.375V Intel's recommended max for 32nm. 4.2GHz at stock voltage is on some processors but not the majority number of processors could do 4.2GHz at stock voltage. Let's just say that I am wrong and Intel really could release a chip at 3.8GHz with no problems whatsoever. Would they? I don't think so because if I take a 4.5GHz as a baseline, we're looking at a 1.1-1.2GHz increase from stock clockspeed which is a huge gain compared to a 0.3-0.7GHz with stock being at 3.8-4.2GHz, the WOW factor isn't that great if Intel increases the stock out of the factory. It could be that Intel has intentionally left quite a lot of headroom just to give their users the perception that SB is a great overclocker.

Again, IB might not be 4GHz but within the 3-4GHz range but much closer to 4GHz than SB. IPC for BD is low because it just how it was made and not because it sucks at clock for clock. It is possible to see on average 5GHz++ @ 1.5V BDs which 1.5V is still a conservative range for BD due to it's claimed robustness. It could possibly go as far as 6GHz easily with a proper custom watercooling, 24/7 stable. A SB although could go till 5.7GHz max, nobody would do that on a 24/7 OC as the voltage required will slowly kill it. What BD can't make up for IPC it gains in clockspeed. IB performance is still unknown and most probably we're comparing Piledriver with IB by then.

Most likely there will not be a 5GHz stock BD in the future. It is likely that some minor improvements is made to BD's architecture to improve its performance. AMD's priority now is not the desktop space and not much effort will be placed in a market they can't compete very well and is not that big anymore. AMD's priority is more of the server space, that's why you see the server BD, Interlagos is released way earlier.

I never said that it is impossible for IB to be 4GHz, I just say that it is more likely that Intel wouldn't. Why would they release the first iteration at 3.8GHz with the current price when they can release a new processor with a 200MHz speed bump over the original iteration and call it an upgrade with a higher price. There is no need for Intel to be pushing the limits of their processors and raise the competitive bar when they are not facing much competition from AMD.