The Really Good Times are Over
We really do not realize how good we had it. Sure, we could apply that to budget surpluses and the time before the rise of global terrorism, but in this case I am talking about the predictable advancement of graphics due to both design expertise and improvements in process technology. Moore’s law has been exceptionally kind to graphics. We can look back and when we plot the course of these graphics companies, they have actually outstripped Moore in terms of transistor density from generation to generation. Most of this is due to better tools and the expertise gained in what is still a fairly new endeavor as compared to CPUs (the first true 3D accelerators were released in the 1993/94 timeframe).
The complexity of a modern 3D chip is truly mind-boggling. To get a good idea of where we came from, we must look back at the first generations of products that we could actually purchase. The original 3Dfx Voodoo Graphics was comprised of a raster chip and a texture chip, each contained approximately 1 million transistors (give or take) and were made on a then available .5 micron process (we shall call it 500 nm from here on out to give a sense of perspective with modern process technology). The chips were clocked between 47 and 50 MHz (though often could be clocked up to 57 MHz by going into the init file and putting in “SET SST_GRXCLK=57”… btw, SST stood for Sellers/Smith/Tarolli, the founders of 3Dfx). This revolutionary graphics card at the time could push out 47 to 50 megapixels and had 4 MB of VRAM and was released in the beginning of 1996.
My first 3D graphics card was the Orchid Righteous 3D. Voodoo Graphics was really the first successful consumer 3D graphics card. Yes, there were others before it, but Voodoo Graphics had the largest impact of them all.
In 1998 3Dfx released the Voodoo 2, and it was a significant jump in complexity from the original. These chips were fabricated on a 350 nm process. There were three chips to each card, one of which was the raster chip and the other two were texture chips. At the top end of the product stack was the 12 MB cards. The raster chip had 4 MB of VRAM available to it while each texture chip had 4 MB of VRAM for texture storage. Not only did this product double performance from the Voodoo Graphics, it was able to run in single card configurations at 800x600 (as compared to the max 640x480 of the Voodoo Graphics). This is the same time as when NVIDIA started to become a very aggressive competitor with the Riva TnT and ATI was about to ship the Rage 128.
Subject: General Tech | April 18, 2013 - 01:46 PM | Ken Addison
Tagged: vsync, vertex 3.20, podcast, pcper, overclocking, ocz, haswell, gtx 780, GLOBALFOUNDRIES, gigabyte brix, frame rating
PC Perspective Podcast #247 - 04/18/2013
Join us this week as we discuss Frame Rating and Vsync, the future of GLOBALFOUNDRIES, the OCZ Vertex 3.20 and more!
The URL for the podcast is: http://pcper.com/podcast - Share with your friends!
- iTunes - Subscribe to the podcast directly through the Store
- RSS - Subscribe through your regular RSS reader
- MP3 - Direct download link to the MP3 file
Hosts: Ryan Shrout, Jeremy Hellstrom, Josh Walrath, and Allyn Malventano
Program length: 1:07:41
0:01:08 Win the Roccat ISKU Keyboard
Week in Review:
News items of interest:
Jeremy: support Full Control not just because they're nordic
Allyn: (portable headsets that don't suck)
1-888-38-PCPER or email@example.com
Taking a Fresh Look at GLOBALFOUNDRIES
It has been a while since we last talked about GLOBALFOUNDRIES, and it is high time to do so. So why the long wait between updates? Well, I think the long and short of it is a lack of execution from their stated roadmaps from around 2009 on. When GF first came on the scene they had a very aggressive roadmap about where their process technology will be and how it will be implemented. I believe that GF first mentioned a working 28 nm process in a early 2011 timeframe. There was a lot of excitement in some corners as people expected next generation GPUs to be available around then using that process node.
Fab 1 is the facility where all 32 nm SOI and most 28 nm HKMG are produced.
Obviously GF did not get that particular process up and running as expected. In fact, they had some real issues getting 32 nm SOI running in a timely manner. Llano was the first product GF produced on that particular node, as well as plenty of test wafers of Bulldozer parts. Both were delayed from when they were initially expected to hit, and both had fabrication issues. Time and money can fix most things when it comes to process technology, and eventually GF was able to solve what issues they had on their end. 32 nm SOI/HKMG is producing like gangbusters. AMD has improved their designs on their end to make things a bit easier as well at GF.
While shoring up the 32 nm process was of extreme importance to GF, it seemingly took resources away from further developing 28 nm and below processes. While work was still being done on these products, the roadmap was far too aggressive for what they were able to accomplish. The hits just kept coming though. AMD cut back on 32nm orders, which had a financial impact on both companies. It was cheaper for AMD to renegotiate the contract and take a penalty rather than order chips that it simply could not sell. GF then had lots of line space open on 32 nm SOI (Dresden) that could not be filled. AMD then voided another contract in which they suffered a larger penalty by opting to potentially utilize a second source for 28 nm HKMG production of their CPUs and APUs. AMD obviously was very uncomfortable about where GF was with their 28 nm process.
During all of this time GF was working to get their Luther Forest FAB 8 up and running. Building a new FAB is no small task. This is a multi-billion dollar endeavor and any new FAB design will have complications. Happily for GF, the development of this FAB has gone along seemingly according to plan. The FAB has achieved every major milestone in construction and deployment. Still, the risks involved with a FAB that could reach around $8 billion+ are immense.
2012 was not exactly the year that GF expected, or hoped for. It was tough on them and their partners. They also had more expenses such as acquiring Chartered back in 2009 and then acquiring the rather significant stake that AMD had in the company in the first place. During this time ATIC has been pumping money into GF to keep it afloat as well as its aspirations at being a major player in the fabrication industry.
Subject: General Tech | February 12, 2013 - 02:10 PM | Jeremy Hellstrom
Tagged: IBM, Samsung, GLOBALFOUNDRIES, CNTFETs, nanotubes
You might not think of IBM, Samsung, and GlobalFoundries as working together for a common goal, but much like the HSA the Common Platform Technology Forum brings together some strange bedfellows. The Tech Report had a chance to sit in on some of the conference and just how this disparate group of Fab owners and pure research companies are working together to shape the future of the silicon beasts we all love to hate. One of the main topics of discussion was the move to the 14nm process and just how designs must change in order to shrink the process to that size while at the same time increasing wafer size, with GloFo showing off their plans for the near future. You will also be introduced to the idea of CNTFETs, the proposed carbon nanotube based replacement for Silicon FinFETs which could beat the limits of even Extreme UV lithography if they can be coerced into self assembly. Read on and check out where the second and third largest Fabs on the planet are headed in the next few years.
"The opportunity doesn't come along every day to get a detailed peek into the future of computing from the people who are building it. Last week, I had just such a chance."
Here is some more Tech News from around the web:
- 'Let anyone be administrator' bug in VMware snapped shut @ The Register
- Remember that Xeon E7-Itanium convergence? FUHGEDDABOUDIT @ The Register
- Google starts rolling out Android 4.2.2 Jelly Bean @ The Inquirer
- Adobe muzzles TWO zero-day wild things with emergency Flash patches @ The Register
- Blackberry uses Samsung Galaxy S3 chips in its Z10 smartphone @ The Inquirer
- Every single Internet Explorer at risk of drive-by hacks until Patch Tuesday @ The Register
- ConnecteDevice COOKOO Watch Review @ Madshrimps
- Netgear R6300 Dual Band Gigabit WiFi Router @ eTeknix
- Unigine Valley & Unigine Heaven 4.0 Coming Next Week @ Phoronix
- Ninjalane Podcast - Console Hardware, HWBot Aquamark Wrapper and EVGA Interview
Subject: General Tech | September 21, 2012 - 01:58 PM | Jeremy Hellstrom
Tagged: 14nm, FinFET, 3d transistors, GLOBALFOUNDRIES, SoC
Intel was first out of the gate with their 3D transistors, which they dubbed Tri-gate and which the rest of the world refers to as FinFET as the normal 2D transistor is flipped on its side in a position reminiscent of a fin. This leads to much more efficient power usage, perfect for mobile designs and needed as the transistor density at 14nm is going to be quite high. GLOFO's 14nm eXtreme Mobility will work in conjunction with the current 20nm process used to fabricate SOCs and will be the basis of many lines of chips, such as ARM who have signed a multiyear contract with GLOFO. Check out DigiTimes for more.
"Globalfoundries has announced the launch of a new technology designed for the expanding mobile market. The new 14nm-XM offering will give customers the performance and power benefits of three-dimensional "FinFET" transistors with less risk and a faster time-to-market, helping the fabless ecosystem maintain its leadership in mobility while enabling a new generation of smart mobile devices, according to the foundry."
Here is some more Tech News from around the web:
- Microsoft releases VMware-EATER @ The Register
- Deep, deep dive inside Intel's next-generation processor @ The Register
- Oh, Sublime Text, how do I love thee @ The Tech Report
- NVIDIA To Publicly Release Some Documentation @ Phoronix
- Logitech C920 HD Pro Webcam Review @ NikKTech (As seen on the PCPer Podcast)
- How to repair a ribbon cable connection on consumer electronics @ Hack a Day
- Apple iOS 6 Review @ TechReviewSource
- Win AFOX HD7850 Single Slot Crossfire @ Kitguru
Subject: Editorial | October 28, 2011 - 05:27 PM | Josh Walrath
Tagged: Q3 2011, ontario, llano, GLOBALFOUNDRIES, bulldozer, brazos, amd
Unlike Intel, AMD was unable to report record revenues. What they were able to report was a small profit. They also were able to show some growth above that expected by most analysts, and even those in AMD. Earlier this quarter AMD warned that revenues might not be as high as expected, but in the end AMD seemed to have done ok.
The company had a gross revenue of $1.69 billion, which is well above the expected $1.66 billion many analysts were predicting. Net profit for the quarter came in at a reasonable $97 million. This is a big improvement from Q3 2010, which had a net income of -$118 million. Being positive for a quarter is a big accomplishment for AMD. Controlling costs as a fabless semiconductor company is a lot easier as compared to running multiple Fabs and researching and implementing next generation process nodes. Margins increased to 45%, but are still a far cry from the 60% plus that Intel achieves. ASPs are also down due to the large amount of low priced, 45 nm parts that AMD still sells.
The primary movers for the positive results for AMD are their lineup of APUs. The “Bobcat” based APUs have been a success for quite a few months, and with their superior performance and features as compared to the competing Intel Atom series, AMD is making a tidy sum off of them. The big winner in the APU sector is of course Llano. The uptake on this processor in the mobile space has been tremendous. AMD has seen a 35% increase in mobile sales, and when combined with the already strong Brazos/Ontario platform, AMD is finally a factor in the mobile market. The only real issue in this market that AMD is facing is that of continued poor yields on Llano processors.
Subject: Editorial | October 12, 2011 - 08:45 AM | Josh Walrath
Tagged: GLOBALFOUNDRIES, fx-8150, bulldozer, am3+, amd, 32 nm
I am pretty sure I am not the only person who has read these Bulldozer reviews (including Ryan's here at PC Perspective) and had that particular reaction. Bulldozer was supposed to bulldoze the competition. It turns out it barely outpaces its own predecessor, the Phenom II X6 1100T. In fact, in terms of IPC, the older Thuban architecture gives it a sound thrashing when both are clocked at 3.3 GHz. So why should I be impressed with this processor?
I guess the answer is… you shouldn’t. At least not yet. I distinctly remember back in November of 2007 being invited to Lake Tahoe to test and report about the first Phenom samples that were available for limited testing. We were not allowed to take the samples home with their new AM2+ based motherboards. When going over the results of the tests with Ryan (I was not part of PCPer at the time) we quickly saw that the 2.6 GHz Phenom was unable to keep up with the Core 2 Q6600 from Intel. This was a little surprising, as we expected the original Phenom to clean house due to its very forward looking architecture (HT, IMC, beefier FP/SIMD units, etc.). The original Phenom had its fair share of problems, to say the least. TDPs were very high, there was the revision B2 bug that was solved in B3, and due to the 65 nm process it did not nearly have as much cache as was needed to make it a more efficient product.
Subject: General Tech | October 3, 2011 - 12:14 PM | Jeremy Hellstrom
Tagged: llano, APU, GLOBALFOUNDRIES, 28nm, 32nm, trinity, amd
Woe is AMD, as it becomes ever more obvious that Llano is not giving good yields at GLOBALFOUNDRIES. Back in July the market noticed that while the new Llano chips were great at providing good enough graphics for a great price, they were very hard to find. As a consumer you might be able to find a notebook to purchase for yourself but as a reseller who needs to buy processors in the thousands before considering that chip as a part of the new product line up you have been out of luck. According to DigiTimes this will change in 2012 with the arrival of Trinity which will still use GLOBALFOUNDRIES 32nm process, turning to TSMC for the 28nm graphical portion. The previous hope that the supply problems would be solved in September were obviously a little too optimistic.
"Supply of AMD's Llano APUs, affected by Globalfoundries's lower-than-expected 32nm yield rates, has been significantly limited and is unlikely to recover until the company's upcoming Trinity arrives in 2012, according to sources from motherboard players. When asked about the company's upcoming Trinity schedule, AMD Taiwan declined to comment on unannounced products.
AMD started suffering from Llano APU supply shortages in July due to the yield issues and the company originally expected the supply status to return to normal in September. However, judging from the current situation, the sources believe the company's supply volume is unlikely to meet client demand through the end of 2011.
The sources estimated that the yield rate issue should be resolved in 2012, when Trinity launches."
Here is some more Tech News from around the web:
- Fanless PCs with 95W CPUs shown off @ SemiAccurate
- Vulnerability in HTC smartphones exposes user data @ The Inquirer
- Mozilla releases Rescuefox prototype @ The Inquirer
- Pandemonium as Microsoft AV nukes Chrome browser @ The Register
- Btrfs File-System For Old Computers? @ Phoronix
- Virtual showdown: Parallels Desktop 7 and VMware Fusion 4 @ Ars Technica
- A Beginner’s Guide to Video Encoding @ t-break
- Real World Labs And Enermax Joint Contest
Subject: General Tech, Processors | September 13, 2011 - 06:07 PM | Ryan Shrout
Tagged: TSMC, idf 2011, idf, GLOBALFOUNDRIES
While learning about the intricacies of the Intel tri-gate 22nm process technology at the Intel Developer Forum, Senior Intel Fellow Mark Bohr surprised me a bit by discussing the competition in the foundry market. Bohr mentioned the performance advantages and competitive edge that the new 22nm technology offers but also decided to mention that other companies like TSMC, GlobalFoundries, Samsung and IBM are behind, and falling further behind as we speak.
When Intel introduced strained silicon in 2003, it took competition until 2007 to implement it. For High-K Metal Gate technology that Intel brought into market in 2007 it wasn't until 2011 that AMD introduced in its product line. Finally, with tri-gate coming in 2011, GlobalFoundries is talking about getting it implemented in the 2015 time frame.
Obviously those are some long delays but more important to note is that the gap between Intel and the field's implementations has been getting longer. Three years for strained silicon, three and a half for high K and up to four years for tri-gate. Of course, we could all be surprised to see tri-gate come from a competitor earlier, but if this schedule stays true, it could mean an increasing advantage for Intel's products over AMD's and eventually into ARM's.
This also discounts the occasional advantage that AMD had over Intel in the past like being the first to integrate copper interconnects (on the first Athlon) and the first to develop a Silicon-on-Insulator product (starting with the 130nm process); though Intel never actually adopted SOI.
Bulldozer Ships for Revenue
Some months back we covered the news that AMD had released its first revenue shipments of Llano. This was a big deal back then, as it was the first 32 nm based product from AMD, and one which could help AMD achieve power and performance parity with Intel in a number of platforms. Llano has gone on to be a decent seller for AMD, and it has had a positive effect on AMD’s marketshare in laptops. Where once AMD was a distant second in overall terms of power and performance in the mobile environment, Llano now allows them to get close to the CPU performance of the Intel processors, achieve much greater performance in graphics workloads, and has matched Intel in overall power consumption.
KY Wong and Marshall Kwait hand off the first box of Bulldozer based Interlagos processors to Cray's Joe Fitzgerald. Photo courtesy of AMD.
Some five months later we are now making the same type of announcement for AMD and their first revenue shipment of the Bulldozer core. The first chips off the line are actually “Interlagos” chips; basically server processors that feature upwards of 16 cores (8 modules, each module containing two integer units and then the shared 256 bit FPU/SSE SIMD unit). The first customer is Cray, purveyor of fine supercomputers everywhere. They will be integrating these new chips into their Cray XE6 supercomputers, which have been purchased by a handful of governmental and education entities around the world.