January 22, 2004

Forums are Back Up! - Josh

It only took 9 months or so, but the forums are back up.  I kept putting this off because I thought we would be on our new design and backend by now, but things just don't always work out!  What can you do?  Anyway, please visit the new and improved forums here.

Happy New Year (Very Late) - Josh

Well, I must admit things have been rather hectic here as of late.  In my real job we switched around most of the computers in the company, built a few more from scratch, and are dealing with some very interesting networking problems.  So that has been taking up a large portion of my time.  The other portion of my time has been spent working on an international adoption!  This past week we were paired with an abandoned Chinese boy from Shanghai, and we should be going to go pick him up sometime this summer.  My wife and I are very excited about adopting this little boy, and he will be named Tai Donald Keith Walrath.  He was born Dec. 17, 2002, so he is just over 13 months old right now.  We can't wait to really start our little family here in Wyoming.

So, now that many of the big explosions in my life are receding, I am working on getting back up and running here!

The Coming Months - Josh

Who says Q1 is boring?  So far we have had Corsair release some really nice, new memory products that should be very welcome to the enthusiasts out there (2 GB matched pair PC3200 DDR DIMMS in both un-buffered and registered form), and of course keep their other high end products in place (with the nice LED's on top).  AMD has released the Athlon 64 3400+, which is one of the fastest processors you can buy.  They also have announced excellent earnings for the last quarter ($1.2 Billion US in revenue and a net income of $42 million US), which should help propel them into the new year.  They are also increasing their ramp of Athlon 64 parts, so we can probably expect around a million of them shipped this current quarter.  With companies like HP and E-machines selling products based on this part, as well as other enthusiast brands like Voodoo PC and Alienware, we can expect some other big names to start picking up the Athlon 64 and integrate it into their lines.  AMD has a solid desktop and mobile roadmap, and are executing on it quite well so far (other than the initial Athlon 64 products being a bit behind schedule).  Intel will be releasing its Prescott based P4 in the next few weeks, and many are quite interested in what that will bring to the table.  An Intel engineer basically just told me that depending on the situation, it could be either 5% faster or slower than a similarly clocked Northwood core P4.  I guess the big thing with the Prescott is the SSE3 and the ability to scale VERY high in clockspeed.  That Intel was able to achieve these design goals and have a part that is still as fast as the current P4 is actually pretty impressive if you think of it.  Adding a bunch more stages to its pipeline, and still having around the same IPC is nothing to scoff at.  We of course have been spoiled by AMD with the Athlon 64 series, which provide the same clockspeeds as the current Athlon XP processors, but have significantly more performance.

On the video side things are about to get interesting again.  When work started on the NV30 and the R300 chips from NVIDIA and ATI respectively, PS/VS 2.0 specifications still were not set in stone.  It appears that NVIDIA took themselves out of the DX9 loop for a significant period of time, during which many of the specifications were put in place.  ATI had the advantage of seeing what was being planned and completed their design with what they thought most likely would be the spec.  NVIDIA came back in pretty late to the party, and there were only so many changes that could be implemented in their design to provide more performance in DX9 apps.  All of that is history now, and NVIDIA has had to live with their basic design, while ATI has been very successful with the R3x0 series of chips.  The DX9 spec, when finalized, also contained PS/VS 3.0 functionality.  This makes PS/VS 3.0 a known quantity for both ATI and NVIDIA when they started design work on the R420 and NV40.  While I can't say who will have the faster or more robust part, I do believe that the playing field will be a lot more level this time.  With ATI's overall experience and success of design for DX9 parts, they could still have an edge here.  However, NVIDIA is not staffed by 2nd class engineers, and there is a lot of pride in that company.  When both the NV40 and R420 parts hit the market, we are going to see a whole new level of performance and feature sets that will be similar to what we saw with the original release of the R300.  I am actually very excited about NVIDIA's new anti-aliasing scheme, which should be a significant step up from what is offered now.  Few details about it are known, only that it will be new and look much better.  About time I say!  ATI's R420 may hold similar surprises, but due to their very successful gamma-corrected jittered multi-sampling scheme, we may not see new AA formats.

On the chipset side for NVIDIA, we can expect to see a DX9 based Athlon 64 product hitting the shelves sometime in Q2 of this year.  Again, very few details about this part have been leaked other than it will probably be a two chip solution.  It will most likely have all the southbridge bells and whistles that we expect (SATA-RAID, Gig-E, etc.) and while it will have integrated sound, there is still a question whether NVIDIA will implement the full featured APU that we saw in the nForce and nForce 2 (MCP-T).  Most likely such a feature will be an option, because it doesn't make any sense not to include that in as a premier solution (especially since it is already designed!).  If NVIDIA does include the full featured APU, we may yet see it become a 24 bit solution.  With Creative, VIA, and others migrating to a 24 bit sound solution, it really makes sense for NVIDIA to follow suite.  With rumors of a standalone audio chip circulating for the past few months, perhaps NVIDIA will only go that route with their high end audio.  Most likely due to the overall bandwidth needs of the APU (which can exceed the bandwidth of the PCI bus), NVIDIA will release this card as a PCI Express product.  Again, this is only speculation.  On the video side of an integrated chipset, there have been no leaks whatsoever about the makeup of the chip.  There are four possible solutions: an integrated version of the FX 5200, an integrated version of the FX 5600, an integrated version of the FX 5700, or it could be new silicon that will be a cut down version of the NV40.  Personally, I would love the latter, but most likely we will see one of the former used due to transistor counts (that would make a shrunk version of the FX 5200 a prime candidate).  I wouldn't be surprised if NVIDIA did produce this part on the 130 nm process at TSMC, or even the 110 nm process (now that it is close to producing revenue products).  Many of the exact details of how NVIDIA will achieve a high performing integrated part are still under wraps, and we won't know the final details until much closer to launch.

All in all, the next few months will be a very interesting time for the computer industry as the technological march moves on.

December 21, 2003

A Small Update and Interesting Read - Josh

Wow, I can't believe it has almost been a month since anything was posted here.  Actually, I can believe it.  Things are of course insane with my real job, as we are doing a major IT audit.  Add to that we moved our entire server room, but oddly enough at the same time Qwest decided to delete our static IP's at the same time (for no particular reason).  So I had to figure out why our internet service didn't work, as well as email, VPN, and ftp.  At first I thought it was something that I did, and after hours of troubleshooting the new server room and setup, I finally decided it wasn't me.  I was correct.  So basically with all of the fun IT stuff, as well as the adoption process we are starting to go through, I have not had a huge amount of time to work on the site.  Keith has been pretty busy as of late also, and Scott is being kept very busy with school and Best Buy.  Hopefully that will change this evening, as Keith has a wireless article that should be posted this evening.  I then will hopefully have an article ready by tomorrow.  And then the day after, we can maybe hope that Scott will have one more article ready!  If so, then it should be a nice read before Christmas begins.

One of my favorite writers, Orson Scott Card, put up an editorial in the Wall Street Journal.  I have a great amount of respect for Mr. Card, and I have always appreciated his writing.  I didn't know this, but he is a Democrat, and very active in that party.  In this article he writes a scathing opinion about the current state of the Democrat party, and some of the things he would like to see come out of it.  I would have to agree with Mr. Card 100%.  Even though I am a Republican, I think that it is key that both parties have strong leaders and candidates, and so far the batch of Presidential hopefulls from the Democrat Party look pretty pathetic.  Mr. Card goes over what he thinks is wrong with the party, the reigning philosophy, and the candidates and their platforms.  A very enlightening read.  I highly recommend that anyone at all interested should read this.

Here's to hoping that we can get a few articles up this week!

November 25, 2003

Back from Comdex - Josh

Well, this partially explains why I haven't updated the site in so long!  I was in Vegas all last week, and this has been my first chance to tell a bit about it.  If you hadn't heard already, this Comdex was very much a disappointment for those looking for a rebirth of the show.  This year only one hall of the LVCC was filled (as compared to 4 years ago where both halls of LVCC and the Sands Convention Center top level were used).  It seems that a more popular thing to do was to use a suite or meeting room in a different hotel, and show your products there.  This involved a lot of walking for Keith and myself, and made for some rather long days.  I did see some interesting things though.  This week I will work on a quick article of what all happened, as well as some of the interesting things that I was able to check out.

One of the more interesting "rumors" that I was able to dig into a bit was the possibility of an Intel based nForce chipset from NVIDIA.  While this speculation has been around for some time, it seems that there is now quite a bit of evidence supporting it.  Apparently Asus is the only motherboard manufacturer with this particular silicon, and had a motherboard working with this product at the show.  When I asked Asus about it, the guy I was talking to got very nervous and started to ask where exactly I had heard this from, and was I from NVIDIA.  I thought this rather odd, but when I said I had merely heard a rumor, he became much more relaxed and laughed a bit, and said there of course was no such thing.  I then asked NVIDIA about it, and all they had to say was that they continue to work closely with Intel and if they get the chance, they would happily create a chipset for the Pentium 4 and the upcoming Prescott.  The other thing I had heard was that Intel was asking for a $4 royalty for each chipset, and that Jen Hsen wanted to only pay half that price.  It seems the two companies still have a long ways to go before they reach an agreement.

The other interesting thing was the amount of motherboard manufacturers that appear to be picking up the ATI Radeon based chipset for the P4.  ATI had just recently released a performance BIOS for this chipset that apparently takes the performance of it very close to the i865 level, and it of course has 1/2 a Radeon 9000 core for the integrated graphics portion.  Overall, this looks to be a very impressive integrated chipset for the Pentium 4, and ATI has finally solved all of its problems with their inhouse Southbridge.  We should expect to see mainstream products based upon this chipset in late December.  This is definitely a huge step up from the Intel Extreme Graphics.

One thing that was not asked, and I am kicking myself for, was the possibility of enhanced versions of the nForce from NVIDIA with integrated NV34 technology.  From all indications NVIDIA is a bit behind on this technology because all of last year they were commenting on how they will offer DX9 from top to bottom (with the bottom being integrated graphics).  While NVIDIA did show off their nForce 3 250 Pro chipset, there were no integrated graphics chips for the Athlon 64.  Of course with the price of the Athlon 64, it doesn't yet make sense to have an integrated part for it.  Not until we see the Athlon 64 reach $200 price points will there really be a need for an integrated chipset.  Perhaps early next year we will see something of this nature.

One other tidbit that I picked up was that chipset sales for NVIDIA, while respectable, were not an overwhelming success for the company.  Many motherboard manufacturers complained that the price of the nForce 2 made it far from a volume product.  The majority of AMD Athlon sales in terms of chipsets were from the VIA KT600.  SiS also sold quite a few of their Athlon based chipsets.  While NVIDIA held a lot of mindshare here, their numbers just were not as impressive as many had expected.

I hope to have more on all this later!

November 2, 2003

Site News and State Article - Josh

Wow, what an unbelievable few days that we have had here.  The editorial has been more than controversial, and for a variety of reasons.  Back when I first started doing articles and reviews, much of the technology behind 3D graphics was a lot more easy to understand.  Pixel pipelines where just that, pixel pipelines.  It was easy to see how things like clock speed, pipeline numbers, memory bandwidth, and other basic principles really affected performance.  The press at the time were a lot more technologically naive, in that it really was permissible to just gloss over the details and not have a complete idea what was going on.  Times have changed.  Many members of the press are a lot more savvy when it comes to the technology behind the chips, and so are users.  This is a good environment for everyone, but it is really hard to find the time and the materials to get a good picture of what is going on.  For the past two years I have really struggled to keep the site going in the face of real life.  While Keith and Scott have lent a hand, the main impetus around here is me.  Add to it all the hack attack last May, and I just haven't had the time to keep up with technology as I should have!  This entire thing has been a great learning experience, and I have met some incredibly helpful people along the way.  Needless to say, I have a lot more information at my fingertips that I can more accurately apply to future articles.

I still believe in many of the things that I wrote in the article, but this is tempered by the new information that I now have.  I will be doing an update to this article, and I will start on the renewed research and writing this week.  I have a couple of other projects that I want to finish in the meantime, so we should have some other articles up and around very soon.

I am slowly rebuilding the site.  I did put up the Editorials and Review indexes again, so those links work from the main page.  I will try to get them working on the other pages if I have the time.  Unfortunately this site is all html, so it is time consuming to go through every page and update the links.  Sorry for the unprofessional appearance of the 404 pages!

Speaking of the site, work is going along well with the new site design and backend.  I was originally hoping to have a beta of the site around mid October, but things of course didn't work out that way.  Now I am hoping to have a working site that I can test in the next three weeks, and a hopeful rollout in early December.  Keep watching, we will keep busy here!

October 30, 2003

The Shocking State of 3D Affairs - Josh

This little article came to me some weeks ago, but it definitely needed a lot of work and research to get it posted.  Many have complained that NVIDIA not only has bad hardware, but do a great amount of cheating to make their products look good.  I decided to take a deeper look at the situation regarding NVIDIA's hardware, software, and "optimizations".  What I found was actually quite enlightening.  There are some very distinct and logical reasons for what NVIDIA has done in the past year in regards to their hardware, software, and the way they have handled the situation.  This is not to say that I agree with all of NVIDIA's tactics, but there some very compelling reasons about why the market is as it is.

I have gone over a bit of history that is not well known around the industry, as well as go over the hardware decisions NVIDIA has made with their GeForce FX series of chips.  Here is a direct quote:

            Since the official release of the GeForce FX 5800, NVIDIA has been under attack by the press and enthusiasts everywhere for their products apparent lack of performance against ATI’s competing Radeon R3x0 series of cards and chips.  Since NVIDIA had such an extensive fan and loyalty base, many users have felt betrayed and let down by the company.  This is not to say that the users who put down $500+ US to buy the thunderous GeForce FX 5800 Ultra should not be mad about the situation, but there is a lot of background not covered by the press that should be brought to light.  There were a series of events that doomed the GeForce FX technology to play second fiddle to ATI’s R300 series in overall performance, and these events happened years ago. 

This should help users understand why NVIDIA's cards do not perform as well as ATI's in current applications, and why NVIDIA may have a distinct advantage when the next generation of hardware hits the streets.  You can read the entire article here.

October 26, 2003

Massive Athlon 64 3200+ and Asus K8V Deluxe Review - Josh

I spent a considerable amount of time with this combination, and have discovered some very interesting things about this particular combination.  Not only have we covered a solid overview of the Athlon 64 3200+ product and architecture, and an exhaustive review of the Asus K8V Deluxe, we have also compared this solution against a slightly older Athlon XP to show how this particular upgrade could help the average user looking for more power.  Pull up a chair, grab something to drink, because this is a big one!  Here is a quote:

     While the FX 51 and Pentium 4 have very similar overall bandwidth numbers, the little Athlon 64 3200+ has only half that number (3.2 GB/sec vs. 6.4 GB/sec). At first glance it would seem that the Athlon 64 3200+ would be at a significant disadvantage, but benchmarks have shown that the Athlon 64 3200+ is at least equal to the Pentium 4 3.2 GHz processor, and only slightly behind the FX 51. Again, this is mostly due to the nature of memory accesses. Streaming data is comparatively uncommon as compared to single or multiple d-words, which is what most applications utilize. Only when a user gets into really heavy duty video editing work, or other such streaming applications, will the Athlon 64 3200+ show its disadvantage. Of course, people who typically do that work have the money to buy the FX 51, which gives the dual advantage of 6.4 GB/sec bandwidth and low latency data accesses.

You can find the entire article here.