Author:
Subject: Editorial
Manufacturer:

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.

righteous3d_01.JPG

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.

Read the entire editorial here!

16nm FinFET ARM processors from TSMC soon

Subject: General Tech | April 2, 2013 - 05:57 PM |
Tagged: arm, FinFET, 16nm, TSMC, Cortex-A57

While what DigiTimes is reporting on is only the first tape out, it is still very interesting to see TSMC hitting 16nm process testing and doing it with the 3D transistor technology we have come to know as FinFET.  It was a 64-bit ARM Cortex-A57 chip that was created using this process, unfortunately we did not get much information about what comprised the chip apart from the slide you can see below.

Cortex_A57_600.jpg

As it can be inferred by the mention that it can run alongside big.LITTLE chips it will not be of the same architecture, nor will it be confined to cellphones.  This does help reinforce TSMC's position in the market for keeping up with the latest fabrication trends and another solid ARM contract will also keep the beancounters occupied.  You can't expect to see these chips immediately but this is a solid step towards an new process being mastered by TSMC.

18198_Cortex-A50-big.LITTLE-sum.jpg

"The achievement is the first milestone in the collaboration between ARM and TSMC to jointly optimize the 64-bit ARMv8 processor series on TSMC FinFET process technologies, the companies said. The pair has teamed up to produce Cortex-A57 processors and libraries to support early customer implementations on 16nm FinFET for ARM-based SoCs."

Here is some more Tech News from around the web:

Tech Talk

Source: DigiTimes

Apple looking to TSMC to replace Samsung as their fab

Subject: General Tech | January 2, 2013 - 01:31 PM |
Tagged: apple, Samsung, TSMC, A6X

According to the news we are hearing from Slashdot and other sources, Apple has finally made the decision to move their chip fabrication from Samsung, for fairly obvious reasons, with TSMC being the lucky fab that will get their business.  They are starting with the current A6X chip, shrinking it to 28nm for the initial run, with the contract for producing the new A7 series dependant on the success of TSMC's trial run.  As the 28nm node is quite familiar to TSMC, barring any production issues that limit availability, they are very likely to win the contract.

burning-bridge.jpg

"The test will kick off in Q1 2013, The China Times reports, with TSMC producing a new, 28nm version of the existing 32nm A6X that Samsung has been producing for the full-sized iPad 4th-gen; the smaller chip, which will likely be more power efficient as well, will debut in a new iPad 5th-gen and iPad mini 2."

Here is some more Tech News from around the web:

Tech Talk

Source: Slashdot

Exclusive partner is another word for single point of failure

Subject: General Tech | August 30, 2012 - 02:38 PM |
Tagged: TSMC, apple, qualcomm, fab

If you believe the rumours, TSMC recently turned down offers from both Apple and Qualcomm to make those companies the exclusive partner of TSMC's smartphone chip production.  Now, that sort of deal does tend to line the pockets of the supplier quite nicely, as the customer must pay to recompense the lost business from other customers.  It also gives the manufacturer the ability to specialize their production lines for one specific type of chip which will eventually bring the cost per wafer down.  On the other hand, this type of deal can stifle innovation on a general level as the manufacturer doesn't need to worry about attracting other customers, nor designing fabrication plants capable of producing multiple types of chips.  Then there is TSMC in specific, a company which has a long history of providing supplies to companies both sides of the war, be it GPU, CPU or a mixed chip.  As arms dealers proved long ago it is far more profitable to sell to both sides than to only supply one belligerent.  Read DigiTimes take on this topic here.

tsmc.jpg

"A recent Bloomberg report cited unnamed sources as saying that Apple and Qualcomm had been rebuffed in separate attempts to invest cash in Taiwan Semiconductor Manufacturing Company (TSMC) in a bid to secure exclusive access to smartphone chips. Digitimes Research analyst Nobunaga Chai has commented saying that he sees no good reason why TSMC should accept the investment."

Here is some more Tech News from around the web:

Tech Talk

Source: DigiTimes

TSMC makes the future of 28nm a little brighter

Subject: General Tech | August 22, 2012 - 03:53 PM |
Tagged: TSMC, 28nm

If you like NVIDIA and AMD's current 28nm process GPUs and future APUs, you should be very happy to hear that TSMC's current 28nm process is topping 80% yield.  This means that the vast majority of silicon coming out of TSMC is good and the supply shortages we have had to become accustomed to are a thing of the past.  This also bodes well for the upcoming AMD APUs which will most likely be using TSMC 28nm silicon.  As well, DigiTimes has good news about the new Fab 15 which should vastly expand TSMC's production capability.

TSMC-1.jpg

"Taiwan Semiconductor Manufacturing Company's (TSMC) 28nm processes have topped a yield rate of 80%, the Chinese-language Commercial Times cited unnamed equipment suppliers as saying in a recent report. Meanwhile, the foundry's new 12-inch fab - Fab 15 - will have a capacity of more than 100,000 12-inch equivalent wafers in the fourth quarter of 2012, according to the report.

Installed capacity at TSMC's Fab 15 thus far in the third quarter has expanded by about 300% sequentially, the report indicated. Fab 15 is identified as the major wafer fab where TSMC's foundry capacity for 28nm processes will be located.

TSMC will be able to satisfy all demand from its major clients including Qualcomm, Nvidia and AMD by the fourth quarter of 2012 as it manages to boost output of wafers processed using 28nm technology, the report believes."

Here is some more Tech News from around the web:

Tech Talk

Source: DigiTimes

ARM, TSMC to Produce 64-bit Processors With 3D Transistors

Subject: Processors | July 24, 2012 - 12:07 PM |
Tagged: TSMC, ARMv8, arm, 64-bit, 3d transistors, 20nm

 

Yesterday ARM announced a multi-year partnership with fab TSMC to produce sub-20nm processors that utilize 3D FinFET transistors. The collaboration and data sharing between the two companies will allow the fabless ARM SoC company the ability to produce physical processors based on its designs and will allow TSMC a platform to further its process nodes and FinFET transistor technology. The first TSMC-produced processors will be based on the ARMv8 architecture and will be 64-bit compatible.

ARMv8.jpg

The addition of 3D transistors will allow the ARM processors to be even more power efficient and suitable for both mobile devices. Alternatively, it could allow for higher clockspeeds at the same TDP ratings as current chips. The other big news is that the chips will be moving to a 64-bit compatible design, which is huge considering ARM processors have traditionally been 32-bit. By moving to 64-bit, ARM is positioning itself for server and workstation adoption, especially with the recent ARM-compatible Windows 8 build due to be released soon. Granted, ARM SoCs have a long way to go before taking market share from Intel and AMD in the desktop and server markets in a big way but it is slowly but surely becoming more competitive with the x86-64 giants.

TSMC’s R&D Vice President Cliff Hou stated that the collaboration between ARM and TSMC will allow TSMC to optimize its FinFET process to target “high speed, low voltage and low leakage.” ARM further qualified that the partnership would give ARM early access to the 3D transistor FinFET process that could help create advanced SoC designs and ramp up volume production.

I think this is a very positive move for ARM, and it should allow them to make much larger inroads into the higher-end computing markets and see higher adoption beyond mobile devices. On the other hand, it is going to depend on TSMC to keep up and get the process down. Considering the issues with creating enough 28nm silicon to meet demand for AMD and NVIDIA’s latest graphics cards, a sub-20nm process may be asking a lot. Here’s hoping that it’s a successful venture for both companies, however.

You can find more information in the full press release.

Source: Maximum PC

AMD and NVIDIA are sticking with TSMC

Subject: General Tech | July 9, 2012 - 04:32 PM |
Tagged: TSMC, Samsung, amd, nvidia, 28nm, rumour

All of the speculation about the problems TSMC has had with their 28nm process and the possible issues they might have producing enough wafers to meet their clients demands.  Today we hear from DigiTimes that Qualcomm is going to switch to Samsung, possibly because TSMC was focusing on AMD and NVIDIA, but this is pure speculation at the moment.  What seems more reliable is that GPU vendors are stating that both AMD and NVIDIA are sticking with TSMC which makes a lot of sense, even if TSMC has problems delivering it is a better alternative than AMD or NVIDIA redesigning their graphics processors to be compatible with Samsung's process.  The story also mentions that in 2013 Brazos 2.0 and Hondo will be moved to a 28nm design, again likely sourced at TSMC.

waffle.jpg

"While Qualcomm has reportedly switched foundry orders for its 28nm-based Snapdragon S4 processors from Taiwan Semiconductor Manufacturing Company (TSMC) to Samsung Electronics because TSMC's 28nm capacity has failed to meet its needs, Nvidia and AMD may not follow suit, according to graphics card makers.

TSMC has the upper hand over Samsung in 28nm technology, yield rate and price and therefore changing foundry partnership involves high risks, the sources said. In addition, Nvidia is expected to consider Samsung's ARM-based processors in competition with its Tegra 3 processors, the sources indicated."

Here is some more Tech News from around the web:

Tech Talk

Source: DigiTimes

Changes afoot for AMD's CPU process

Subject: General Tech | June 18, 2012 - 06:21 PM |
Tagged: amd, TSMC, 28nm

It looks like AMD is going to move its CPU process from the current Silicon on Insulator to the same 28nm bulk CMOS process at TSMC that they currently are using for their Southern Islands GPUs.  This should see the same changes to thermals as with the HD7xxx series of cards when compared to their predecessors and it is possible that this move could help future APUs as both GPU and CPU portions will be using the same improved process.  The article on DigiTimes also carries a bit of speculation on Sea Island, the next GPU architecture as well as the HSA and ARM on AMD.

wafer.jpg

"AMD is set to make a major change in its manufacturing process in 2013 and will fully switch from the existing SOI manufacturing processor to 28nm Bulk CMOS process, according to Mark Papermaster, senior vice president and chief technology officer of AMD."

Here is some more Tech News from around the web:

Tech Talk

 

Source: DigiTimes

Good news from TSMC for NVIDIA and you

Subject: General Tech | May 9, 2012 - 02:28 PM |
Tagged: TSMC, 28nm, nvidia

In the seemingly never ending saga of TSMC, NVIDIA and the mysteriously lacking supply of 28nm GPUs on the market is another update from DigiTimes.  TSMC is going to give priority to NVIDIA on their production lines, though if TSMC is still at 95% capacity that may not mean a great increase in capacity.  This doesn't refute the rumours that NVIDIA is shopping around for a new supplier for their chips nor that they may be revamping the mask they use for the chips but it does imply that TSMC does not want to lose NVIDIA's business and might have some capacity to spare for them.

out_of_stock_main.png

"Since Nvidia has been unsatisfied with TSMC's 28nm process, while the company has also not refuted rumors that the company may cooperate with Samsung Electronics or Globalfoundries, TSMC, to sooth Nvidia, has put the GPU maker on its supply priority, allowing Nvidia to be able to release its 28nm GPUs on schedule in May and June."

Here is some more Tech News from around the web:

Tech Talk

 

Source: DigiTimes

May it truly be the end of our graphics card drought

Subject: General Tech | May 3, 2012 - 12:21 PM |
Tagged: NVIDA, amd, graphics card, TSMC, 28nm, kepler, southern islands, gtx690, gtx680, gtx670, gtx610

Enthusiasts were offered a bit of hope this morning with news from DigiTimes that more capacity at TSMC will be available for AMD and NVIDIA which would mean more dies being made and hopefully a larger supply of GPUs.  Since TSMC seems to have finished playing with their Cortex A9 process, there is a good possibility that the GTX680 and perhaps even the GTX690 will become common enough that the great unwashed actually have a chance to purchase one.  We can also hope that it will give NVIDIA a chance to build up stocks of the GTX670 and 610 which are due out at the end of the month and June, respectively.  Unfortunately, if a certain site is correct that may not be the case as NVIDIA will be redoing their mask and not be able to take advantage of the extra capacity TSMC could make available for them.  Perhaps if this scenario is true AMD will be able to leverage TSMC to flood the market with Southern Island GPUs and hope to win the availability war as the performance crown is firmly on NVIDIA's head in this generation of GPUs.

happydance.png

"AMD and Nvidia, impacted by Taiwan Semiconductor Manufacturing Company's (TSMC's) fully-booked capacity, had rather weak shipment performance in the first quarter; however, as more capacity will be gradually released by TSMC, shortages of 28nm graphics cards are expected to improve in late May, according to sources from graphics card makers."

Here is some more Tech News from around the web:

Tech Talk

 

Source: DigiTimes