Bulldozer to Vishera
Bulldozer is the word. Ok, perhaps it is not “the” word, but it is “a” word. When AMD let that little codename slip some years back, AMD enthusiasts and tech journalists started to salivate about the possibilities. Here was a unique and very new architecture that promised excellent single thread performance and outstanding multi-threaded performance all in a package that was easy to swallow and digest. Probiotics for the PC. Some could argue that the end product for Bulldozer and probiotics are the same, but I am not overly fond of writing articles containing four letter colorful metaphors.
The long and short of Bulldozer is that it was a product that was pushed out too fast, it had specifications that were too aggressive for the time, and it never delivered on the promise of the architecture. Logically there are some very good reasons behind the architecture, but implementing these ideas into a successful product is another story altogether. The chip was never able to reach the GHz range it was supposed to and stay within reasonable TDP limits. To get the chip out in a timely manner, timings had to be loosened internally so the chip could even run. Performance per clock was pretty dismal, and the top end FX-8150 was only marginally faster than the previous top end Phenom II X6 1100T. In some cases, the X6 was still faster and a more competent “all around” processor.
There really was not a whole lot for AMD to do about the situation. It had to have a new product, and it just did not turn out as nicely as they had hoped. The reasons for this are legion, but simply put AMD is competing with a company that is over ten times the size, with the resulting R&D budgets that such a size (and margins) can afford. Engineers looking for work are a dime a dozen, and Intel can hire as many as they need. So, instead of respinning Bulldozer ad nauseum and releasing new speed grades throughout the year by tweaking the process and metal layer design, AMD let the product line sit and stagnate at the top end for a year (though they did release higher TDP models based on the dual module FX-4000 and triple module FX-6000 series). Engineers were pushed into more forward looking projects. One of these is Vishera.
Subject: Processors | October 2, 2012 - 04:56 PM | Jeremy Hellstrom
Tagged: vishera, trinity, Steamroller, piledriver, bulldozer, amd, a8, a6, A4, a10, 5800K, 5600K
The NDA is over and we can finally tell you all about the new generation of Trinity, especially the compute portion which we were not allowed to discuss in the controversial preview. Part of the good news is the price, Legit Reviews found the highest MSRP is $122 for the A10-5800K and it is currently available, though at $130. The performance increase from the previous generation is decent for multicore applications though not so much for single threaded applications, overall you can expect general computing performance in line with Core i3 but not Core i5. Gaming on the other hand did show much improvement, especially with you compare the built in HD7660D to Intel's current HD4000 and HD3500. You can catch Josh's review right here.
"The internal testing from AMD that we can see above shows a 37% increase in the 3DMark 11 score between the first generation A-Series Llano and this generation of A-Series Trinity. While our numbers don't match their numbers exactly, our Llano system scored 1115 3Dmarks while the AMD internal testing showed 1150 3DMarks. Our AMD A10-5800K scored 1521 3DMarks while they scored 1570. The overall difference was remarkably similar, AMD is boasting an increase of 37% and we saw a difference of 36.4%..."
Here are some more Processor articles from around the web:
- AMD’s Trinity Faces Off With Intel’s Ivy Bridge @ SemiAccurate
- AMD “Virgo” Platform: 2nd Generation APU @ Bjorn3D
- AMD A10-5800K APU Performance Review @ HardwareHeaven
- AMD A10-5800K and A8-5600K APUs for Socket FM2 @ techPowerUp
- AMD A10-5800K Trinity APU Review @ TechwareLabs
- Asus F2A85-V Pro & AMD A10 5800K (w/ HD7660D) @ Kitguru
- AMD A10-5800K & A8-5600K Review: Trinity on the Desktop, Part 2 @ AnandTech
- AMD A10 5800K APU processor review and MSI FM-2 A85XA-G65 @ Guru of 3D
- AMD A10-5800K Unlocked "Trinity" Quad Core APU Review @ Hi Tech Legion
- AMD A8-3850 CPU review @ Rbmods
- Gigabyte F2A85X-UP4 & AMD A10 5800K @ Kitguru
- AMD A10-5800K / A8-5600K full review: Trinity for desktops @ Hardware.info
- AMD Trinity for Desktops. Part 1: Graphics Core @ X-bit Labs
- Workstation & Server CPU Comparison Guide @ TechARP
- All Core i3 Models @ Hardware Secrets
- Intel Core i3 3225 and 3220 review: entry-level Ivy Bridge @ Hardware.info
Trinity Finally Comes to the Desktop
Trinity. Where to start? I find myself asking that question, as the road to this release is somewhat tortuous. Trinity, as a product code name, came around in early 2011. The first working silicon was shown that Summer. The first actual release of product was the mobile part in late Spring of this year. Throughout the summer notebook designs based on Trinity started to trickle out. Today we cover the release of the desktop versions of this product.
AMD has certainly had its ups and downs when it comes to APU releases. Their first real APU was Zacate, based on the new Bobcat CPU architecture. This product was an unmitigated success for AMD. Llano, on the other hand, had a pretty rocky start. Production and various supply issues caused it to be far less of a success than hoped. These issues were oddly enough not cleared up until late Spring of this year. By then mobile Trinity was out and people were looking towards the desktop version of the chip. AMD saw the situation, and the massive supply of Llano chips that it had, and decided to delay introduction of desktop Trinity until a later date.
To say that expectations for Trinity are high is an understatement. AMD has been on the ropes for quite a few years in terms of CPU performance. While the Phenom II series were at least competitive with the Core 2 Duo and Quad chips, they did not match up well against the latest i7/i5/i3 series of parts. Bulldozer was supposed to erase the processor advantage Intel had, but it came out of the oven as a seemingly half baked part. Piledriver was designed to succeed Bulldozer, and is supposed to shore up the architecture to make it more competitive. Piledriver is the basis of Trinity. Piledriver does sport significant improvements in clockspeed, power consumption, and IPC (instructions per clock). People are hopeful that Trinity would be able to match the performance of current Ivy Bridge processors from Intel, or at least get close.
So does it match Intel? In ways, I suppose. How much better is it than Bulldozer? That particular answer is actually a bit surprising. Is it really that much of a step above Llano? Yet another somewhat surprising answer for that particular question. Make no mistake, Trinity for desktop is a major launch for AMD, and their continued existence as a CPU manufacturer depends heavily on this part.
Subject: Processors | September 29, 2012 - 10:46 PM | Tim Verry
Tagged: vishera, piledriver, amd, am3+
Trinity APUs are not the only Piledriver-based processors that AMD will be releasing this year. Trinity is coming next month, but later this year AMD should be putting out Vishera processors based on Piledriver CPUs cores – and without integrated GPUs. And now, thanks to a retailer leaking details on its website, we now know some basic specifications – and more importantly – pricing.
For the uninitiated, Vishera is AMD’s next generation processor. It will use the existing AM3+ socket, and is built on a 32nm HKMG manufacturing process. Further, the CPUs are based on the Piledriver architecture which features a number of efficiency improvements over Bulldozer. Thanks to the architecture tweaks, and Cyclos Semiconductor’s resonant clock mesh technology that reduces the amount of power needed to keep the clock frequency synced across the entire chip. The architecture tweaks result in improved instructions per clock (IPC), improved floating point performance, leakage reduction, AMD Turbo Core 3, and new FMA3, AVX, AVS1.1, AES, and F16C instructions among other improvements.
For more information on the Piledriver architecture, and where AMD is taking it with Vishera, read the “AMD: Vishera and Beyond” editorial we recently posted. Also relevant is our mobile Trinity (A10-4600M) review which gives some small hints at the kind of CPU improvements we can expect with desktop Piledriver CPU cores versus the previous generation.
According to eTeknix, the recently leaked information from Bottom Line Telecomunications includes clock speed, core count, amount of cache, TDP and pricing for four of AMD's upcoming FX series Vishera processors: the FX 4300, FX 6300, FX 8320, and FX 8350. The FX 4300 is a quad core processor clocked at 3.8GHz with 8MB of cache and a 95W TDP (thermal design power). It was priced at $131.62 on the company's website. The FX 6300 CPU brings the core count up to six, and increases the cache to 14MB. It keeps the same 95W TDP as the FX 4300 but is clocked at 3.5GHz and costs $175.77.
The FX 8320 and FX 8350 are both eight core processors and have a 125W TDP. The FX 8320 is a $242.05 part with 16MB cache and comes clocked at 3.5GHz. The FX 8350 keeps the same 16MB cache but is clocked at 4GHz and, as a result, costs more at $253.06.
The FX 8320 in particular appears to be a neat processor, and will likely be the more popular of the two FX 8000 series as enthusiasts will overclock it match (or exceed) the FX 8350 while paying the cheaper price (since the only thing you are really giving up with the lower-end part is clockspeed, and not cache)!
It will be interesting to see if the Piledriver-based chips are worth the price though, since we have yet to see independant CPU performance benchmarks for either Vishera or Trinity. The following table is the leaked information from shopBLT mentioned above in table form.
|shopBLT Item #||Manufacturer Part #||Description||Price|
|BPW4489||FD4300WMHKBOX||FX 4300 QC CPU AM3+ 8MB 95W 3.8GHz Box||$131.62|
|BPW4488||FD6300WMHKBOX||FX 6300 6C CPU AM3+ 14MB 95W 3.5GHz Box||$175.77|
|BPW4487||FD8320FRHKBOX||FX 8320 8C CPU AM3+ 16MB 125W 3.5GHz box||$242.05|
|BPW4486||FD8350FRHKBOX||FX 8350 8C CPU AM3+ 16MB 125W 4GHz Box||$253.06|
Speaking of pricing, AMD will not only be competing with Intel's Sandy Bridge processors, but its latest Ivy Bridge chips as well, so pricing will be key to AMD selling its CPUs. In the following chart, we compared AMD's upcoming Vishera processors (based on the leaked information above) to Intel's latest Ivy Bridge parts. Because we do not know what the performancer of Piledriver will be, we matched up the Bulldozer CPUs to the Intel competition based on pricing. Essentially, we attempted to find the the Ivy Bridge CPU with the closest price tag to the Vishera processors' price. Intel's 22nm process has definitely given the company a leg up on TDPs, but you do get as many as twice the cores (and cache) with AMD for the price. The FX 8350 is an odd part in that it does not have a good Ivy Bridge equivalent, because there is no approximately $250 Ivy Bridge CPU. The next-closest CPU is the Core i7-3770 at just-over $300. Note that it may end up being that a lower priced chip will actually perform equivalently (or outperform) to the FX 8350 – we just do not know at this point and the only basis for matching these up for sake of comparison is price right now.
|Processor Model||FX 4300||FX 6300||FX 8320||FX 8350||Core i3 3220||Core i5 3550P||Core i5-3570K||Core i7 3770|
|No. of cores (HT)||4||6||8||8||2 (4)||4||4||4 (8)|
|Clockspeed (turbo)||3.8GHz||3.5GHz||3.5GHz||4GHz||3.3GHz||3.1GHz (3.5)||3.4GHz (3.8)||3.4GHz (3.9)|
The Intel processors were chosen base on pricing and not performance per-se. Note that the i5-3550P does not include integrated graphics.
Another interesting match up is the comparison between AMD's next generation Vishera processors and its current generation Zambezi Bulldozer CPUs.
The FX 4300 cache number seems like the only oddity, but is based on leaked information above.
Assuming that the leaked pricing ends up being accurate, AMD has put itself in an odd position with Vishera. Across the board, the Piledriver-based chips are notably more expensive than the Bulldozer predecessors. The next generation chips are offering up higher clockspeeds – and in some cases – lower TDPs. On the other hand, they are coming in at a premium, and AMD is already facing stiff competition from Intel’s Ivy Bridge and Sandy Bridge processors.
AMD will really have to bring the promised performance improvements in order to move its Vishera chips at these prices. Performance is key, and unfortunately that's one aspect of Piledriver that we don't yet know beyond AMD's claims. Personally, I'm hopeful that they will deliver on the claimed efficiency tweaks and that Vishera will be a success. At the very least, it should offer a nice upgrade for owners of AM3+ motherboards.
After the Trinity launch, we should have more information on the the level of CPU performance we can expect from Piledriver. Keep an eye on PC Perspective for more information on Vishera and the Piledriver architecture in general as it comes in!
Read more about AMD's Piledriver microarchitecture.
Trinity's GPU Performance
Editor's Note: Right before the release of this story some discussion has been ongoing at other hardware sites about the methods AMD employed with this NDA and release of information. Essentially, AMD allowed us to write about only the gaming benchmarks and specifications for the Trinity APU, rather than allowing the full gamut of results including CPU tests, power consumption, etc. Why? Obviously AMD wants to see a good message be released about their product; by release info in stages they can at least allow a brief window for that.
Does it suck that they did this? Yes. Do I feel like we should have NOT published this because of those circumstances? Not at all. Information is information and we felt that getting it to you as soon as possible was beneficial. Also, because the parts are not on sale today we are not risking adversely affecting your purchasing decision with these limited benchmarks. When the parts DO go on sale, you will have our full review with all the positives and negatives laid out before you, in the open.
This kind of stuff happens often in our world - NVIDIA sent out GTX 660 cards but not GTX 650s because of lack luster performance for example - and we balance it and judge it on a case by case basis. I don't think anyone looking at this story sees a "full review" and would think to make a final decision about ANY product from it. That's not the goal. But just as we sometimes show you rumored specs and performance numbers on upcoming parts before the NDAs expire, we did this today with Trinity - it just so happens it was with AMD's blessing.
AMD has graciously allowed us the chance to give readers a small glimpse at the performance of the upcoming A series APUs based on the Trinity processor. Today we are covering the SKUs that will be released, general gaming performance, and what kind of power consumption we are seeing as compared to the previous Llano processor and any Intel processor we can lay hands upon.
Trinity is based on the updated Piledriver architecture, which is an update to Bulldozer. Piledriver improves upon IPC by a small amount over Bulldozer, but the biggest impact is that of power consumption and higher clockspeeds. It was pretty well known that Bulldozer did not hit the performance expectations of both AMD and consumers. Part of this was due to the design pulling more power at the target clockspeeds than was expected. To remedy this, AMD lowered clockspeeds. Piledriver fixes most of those power issues, as well as sprinkles some extra efficiency into the design, so that clockspeeds can scale to speeds that will make these products more competitive with current Intel offerings.
The top end model that AMD will be offering of the socket FM2 processors (for the time being) is the A10 5800K. This little number is a dual module/quad core processor running at 3.8 GHz with a turbo speed of 4.2 GHz. We see below the exact model range of products that AMD will be offering. This does not include the rumored Athlon II editions that will have a disabled GPU onboard. Each module features 2 MB of L2 cache, for a total of 4 MB on the processor. The A10 series does not feature a dedicated L3 cache as the FX processors do. This particular part is unlocked as well, so expect some decent overclocking right off the bat.
The A10 5800K features the VLIW 4 based graphics portion, which is significantly more efficient than the previous VLIW 5 based unit in Llano (A8 3870K and brethren). Even though it features the same number of stream processors as the 3870K, AMD is confident that this particular unit is upwards of 20% faster than the previous model. This GPU portion is running at a brisk 800 MHz. The GPU core is also unlocked, so expect some significant leaps in that piece of the puzzle as well.
That is about all I can give out at this time, since this is primarily based on what we see in the diagram and what we have learned from the previous Trinity release (for notebooks).
Subject: Motherboards | September 17, 2012 - 11:09 PM | Josh Walrath
Tagged: trinity, piledriver, fm2, amd, A85X, a10
Gigabyte lit the social media fuse and showed off some of the first pictures of one of the A85X based motherboards. A85X is the successor to the original FM1 A75 chipset, and it had a rather robust featureset for a "budget" oriented chipset. The original A75 was paired with the Llano APU, otherwise known as the A8/A6/A4 APU from AMD. The A85 is pin compatible with the A75, but it offers two more SATA 6 ports than the previous unit. Both share 14 USB ports, four of which are USB 3.0
The board overall looks nice and robust. The black PCB and accoutrements make it seem like it is a mean board. There are 4 USB 3.0 ports on the back and a header for front panel USB 3.0. All eight SATA 6 ports are used on the board, six + one on the board and one e-SATA. We do not know all the details about the power delivery system, but it looks like it is using a variant of what we saw with the latest Z77 boards from Gigabyte. Good stuff, Mainerd.
October certainly looks to be the month that Trinity arrives. Everyone is very curious how it will perform against the latest Ivy Bridge processors from Intel. While AMD still has a GPU advantage, it is slowly shrinking. Now we wonder how well the CPU part will perform and how much power it will pull. Stay tuned, gentle readers...
Subject: General Tech | September 11, 2012 - 01:47 PM | Jeremy Hellstrom
Tagged: seamicro, amd, Intel, xeon, piledriver, smug
To think that only 3 years ago we finally saw the end of the legal battle between Intel and AMD over the x86 patent makes today's news bring a smile to those with a certain sense of humour. Some of SeaMicro's new servers will be powered by Intel's Xeon line of processors, meaning that an AMD owned company will be offering Intel Inside. As AMD purchased SeaMicro for their "Freedom" 3D mesh/torus interconnect technology as opposed to an attempt to push Intel out of that particular make of server, this move makes perfect sense as AMD's bottom line will benefit from every sale of an Intel based SeaMicro server. It also opens up the choices available to the market as you will be able to purchase Piledriver based SeaMicro servers using the same interconnect technology.
From The Register we get more information on the Piledriver processors we will see in these servers, they will have eight cores and would come in three speeds; 2GHz, 2.3GHz, and 2.8GHz. They also infer that with this design you could have 512 cores and 4TB of memory in a 10U chassis which is enough to make any SETI@Home or Folding@Home team member drool with jealousy. On the Intel side they will use the 2.5GHz quad core Xeon E3-1265L v2 which means you would only have a mere 256 cores in a similar 10U chassis. DigiTimes also picked up on this story with more details on the insides of the servers, both Intel and AMD.
"SeaMicro is not longer an independent company, but you would not have guessed that if you were dropped in from outer space to attend the launch of the new SM15000 microserver in San Francisco on Monday afternoon. Advanced Micro Devices may own SeaMicro, but the company went out of its way to support the latest "Ivy Bridge" Xeon E3-1200 v2 processor from rival Intel as well as its own forthcoming "Piledriver" Opteron processor as new compute nodes in a new SeaMicro chassis."
Here is some more Tech News from around the web:
- Intel hints at weaving network fabric into Xeons, Atoms @ The Register
- Microsoft to open 32 pop-up retail stores for the holidays @ The Register
- NETGEAR N600 Wireless Dual Band Gigabit ADSL 2+ Modem Router Review @ HardwareHeaven
- Samsung will seek to ban Apple's Iphone 5 @ The Inquirer
- A preview of the reviews for the iPhone 5 @ The Tech Report
Subject: General Tech | September 6, 2012 - 02:58 PM | Jeremy Hellstrom
Tagged: vishera, trinity, Steamroller, piledriver, hot chips, bulldozer, amd, Abu Dhabi
You've seen the slides everywhere and read through what Josh could observe and predict from those slides but at the end of Hot Chips will still know little more about the core everyone is waiting for. The slides show a core little changed from Bulldozer, which is exactly what we've been expecting as AMD has always described Steamroller as a refined Bulldozer design, improving the existing architecture as opposed to a complete redesign. SemiAccurate did pull out one little gem which might mean good news for both AMD and consumers which pertains to the high density libraries slide. The 30% decrease in size and power consumption seems to have been implemented by simply using the high density libraries that AMD uses for GPUs. As this library already exists, AMD didn't need to spend money to develop it, they essentially managed this 30% improvement with a button press, as SemiAccurate put it. This could well mean that Steamroller will either come out at a comparatively low price or will give AMD higher profit margins ... or a mix of both.
"With that in mind, the HDL slide was rather interesting. AMD is claiming that if you rebuild Bulldozer with an HDL library, the resulting chip has a 30% decrease in size and power use. To AMD at least, this is worth a full shrink, but we only buy that claim if it is 30% smaller and 30% less power hungry, not 30% in aggregate. That said, it is a massive gain with just a button press.
AMD should be applauded, or it would have been, but during the keynote, the one thing that kept going through my mind was, “Why didn’t they do this 5 years ago?”. If you can get 30% from changing out a library to the ones you build your GPUs with, didn’t someone test this out before you decided on layout tools?"
Here is some more Tech News from around the web:
- The best and worst of IFA 2012 @ The Inquirer
- US energy lab's pump-happy petaflopper goes green @ The Register
- Quantum Teleportation Sends Information 143 Kilometers @ Slashdot
- Finger-free Kinect coming to fondlesome Windows 8 @ The Register
- VIA suffers close to 15% on-month drop in August revenues @ DigiTimes
- Micron expresses interest in partnering with TSMC @ DigiTimes
- An Argument Against Expensive Solid State Drives @ Benchmark Reviews
- Interview: 2Dawn Games on its upcoming shooter 'Ravaged' and life as an indie studio @ TechSpot
Ah, the end of August. School is about to start. American college football is about to get underway. Hot Chips is now in full swing. I guess the end of August caters to all sorts of people. For the people who are most interested in Hot Chips, the amount of information on next generation CPU architectures is something to really look forward to. AMD is taking this opportunity to give us a few tantalizing bits of information about their next generation Steamroller core which will be introduced with the codenamed “Kaveri” APU due out in 2013.
AMD is seemingly on the brink of releasing the latest architectural update with Vishera. This is a Piledriver+ based CPU that will find its way into AM3+ sockets. On the server side it is expected that the Abu Dhabi processors will also be released in a late September timeframe. Trinity was the first example of a Piledriver based product, and it showed markedly improved thermals as compared to previous Bulldozer based products, and featured a nice little bump in IPC in both single and multi-threaded applications. Vishera and Abu Dhabi look to be Piledriver+, which essentially means that there are a few more tweaks in the design that *should* allow it to go faster per clock than Trinity. There have been a few performance leaks so far, but nothing that has been concrete (or has shown final production-ready silicon).
Until that time when Vishera and its ilk are released, AMD is teasing us with some Steamroller information. This presentation is featured at Hotchips today (August 28). It is a very general overview of improvements, but very few details about how AMD is achieving increased performance with this next gen architecture are given. So with that, I will dive into what information we have.
Less Risk, Faster Product Development and Introduction
There have been quite a few articles lately about the upcoming Bulldozer refresh from AMD, but a lot of the information that they have posted is not new. I have put together a few things that seem to have escaped a lot of these articles, and shine a light on what I consider the most important aspects of these upcoming releases. The positive thing that most of these articles have achieved is increasing interest in AMD’s upcoming products, and what they might do for that company and the industry in general.
The original FX-8150 hopefully will only be a slightly embarrasing memory for AMD come Q3/Q4 of this year.
The current Bulldozer architecture that powers the AMD FX series of processors is not exactly an optimal solution. It works, and seems to do fine, but it does not surpass the performance of the previous generation Phenom II X6 series of chips in any meaningful way. Let us not mention how it compares to Intel’s Sandy Bridge and Ivy Bridge products. It is not that the design is inherently flawed or bad, but rather that it was a unique avenue of thought that was not completely optimized. The train of thought is that AMD seems to have given up on the high single threaded performance that Intel has excelled at for some time. Instead they are going for good single threaded performance, and outstanding multi-threaded performance. To achieve this they had to rethink how to essentially make the processor as wide as possible, keep the die size and TDP down to reasonable sizes, and still achieve a decent amount of performance in single threaded applications.
Bulldozer was meant to address this idea, and its success is debatable. The processor works, it shows up as an eight logical core processor, and it seems to scale well with multi-threading. The problem, as stated before, is that it does not perform like a next generation part. In fact, it is often compared to Intel’s Prescott, which was a larger chip on a smaller process than the previous Northwood processor, but did not outperform the earlier part in any meaningful way (except in heat production). The difference between Intel and AMD in this aspect is that as compared to Prescott, Bulldozer as an entirely new architecture as compared to the Prescott/Northwood lineage. AMD has radically changed the way it designs processors. Taking some lessons from the graphics arm of the company and their successful Radeon brand, AMD is applying that train of thought to processors.