Please create an account to participate in the Slashdot moderation system

 



Forgot your password?
typodupeerror
×
Technology IT

VIA's New PT Chipsets 91

TheTechLounge writes "Today VIA is announcing their new PT series of chipsets to the masses. The chipsets that make up the PT series represent the first real alternatives to Intel's chipsets for the Pentium 4 platform and aim to ease the transition to PCI-Express and DDR-II. All of VIA's PT products are covered under a ten-year cross license agreement between VIA and Intel. As expected, the majority of motherboard manufacturers will be using the PT chipsets in upcoming boards. Some of these companies include Abit, Asus, Chaintech, Biostar, DFI, EPoX, Gigabyte, MSI and Soltek. The PT chipsets cover a wide range of PCI-Express, AGP and IGP solutions for the Intel platform. VIA's new PT chipsets include the PT880 Pro, the PT894 and the PT894 Pro."
This discussion has been archived. No new comments can be posted.

VIA's New PT Chipsets

Comments Filter:
  • Yeah, but... (Score:5, Insightful)

    by Goronmon ( 652094 ) * on Monday January 31, 2005 @02:26PM (#11530425)
    As for pricing, the PT880 Pro will be priced competitively with the current Intel 865 solutions on the market while the PT894 will be priced to compete with the current Intel 915 boards and the PT894 Pro competing with 915/925 boards.

    The fact that they don't mention price until the end and in such a lackluster way it makes it tough for me to get excited about this. I really would like to see a less expensive alternative to Intel, not just "priced competitively".

    Plus, the fact that the benchmarks don't show anything too exciting doesn't help either.

    But competition is always a good thing, I just wish the only selling point didn't seem like "We aren't Intel".
    • by mmkkbb ( 816035 ) on Monday January 31, 2005 @02:33PM (#11530509) Homepage Journal
      and now you understand the woe of the Democrat.
    • I agree. Those benchmarks show marginal improvements at best, and in some cases slight declines in performance, so the only real point these guys seem to be able to compete on is price. It's amazing that the bulk of the article talks about performance, which they have no compelling advantage with, and almost none of it on price.

      But then, since they are forced to license a lot of their technology from Intel, it may not be feasible for them to offer these things at a price substantially lower than Intel's
      • I suppose the thing that really surprises me is that in an article that is clearly a marketing press release, no really compelling reason to use these instead of Intel's products is offered.

        Yeah, I definately got the same "marketing BS" feeling when reading the "article." Which like I said, makes it even more unattractive since you would think with such an article they could come up with better things to say than it just being comparable to Intel.
      • Re:Yeah, but... (Score:3, Insightful)

        by Grand ( 152636 )
        Yea the benchmarks are about the same as Intels. But the thing that makes this special is that you can use 2 different types of ram (not at the same time of course) and two different types of video cards (same time im assuming, I just looked at the pretty pictures in the article). That makes this board much more appealing to people in the MB/upgrading market. Buy this and buy the DDR and AGP video card relatively cheaply. In most cases, people already have these items in their current computer and they o
        • But again, thats only good for the transition phase between. Once/If DDR-II and PCI-E become the standard, that "advantage" goes out the window.
      • Their Intel chipset business has been very bad from a profit perspective for a long time while their small form factor Epia line has been doing quite well and has been a bright spot of cooperation between FOSS software and a major hardware producer. The problem is that the latter is a tiny fraction of the former in terms of revenues because everything is marketed in terms of gaming performance these days despite the fact that a huge segment of PC users don't play cutting edge games.
        At least in part
    • by LWATCDR ( 28044 ) on Monday January 31, 2005 @02:39PM (#11530562) Homepage Journal
      At least now that Intel has gotten over that whole RAMBUS stupidity.
      Intel chip sets tend to be very stable. I have to admit that for a server I was thinking of building I am thinking very hard about an Intel motherboard with an Intel CPU. Unless the VIA is faster or cheaper what is the benefit?
      • Yeah, I agree. Intel just doesn't have the same stigma of some other companies like Microsoft (yeah, I know, MS is used for such comparisons too often, I took the easy choice.)
        • by LWATCDR ( 28044 ) on Monday January 31, 2005 @03:21PM (#11530955) Homepage Journal
          Intel does not have a stigma because for years and years their was little choice but Intel. Even AMD where just faster or cheaper versions of Intel chips until the Athlon64. I think AMD is missing the boat by not making Athlon/Opteron motherboards. It would help eliminate finger pointing. It is the CPU, not it is the chip set, no it is the mother board. If you buy a motherboard from Asus for the Opteron using a Via chip set you now have no less than three companies involved in one major part of your system. You can get an Intel motherboard that uses an Intel chip set with an Intel CPU then you only have one place to look for answers. Of course they will probably blame the ram. All in all Intel has the reputation for very stable motherboards.

          • If you buy a motherboard from Asus for the Opteron using a Via chip set you now have no less than three companies involved in one major part of your system.

            And what's wrong with that? That's exactly the combination that I have in my computer, and it works damn fine. Why does it matter who builds and designs the parts, as long as they work well together?
            • Nothing is wrong if they do work well together. There is a big difference between a server and a desktop when it comes to stability. If a desktop crashes it only effects one person if a server crashes it can effect hundreds or thousands. The last server I built has been up for around five years now. I used a PII and an ABIT motherboard, 3dware raid controller, and Suse. For it's replacement I am thinking Opteron or Xeon. The problem is that I do not have has much time to play around with it as I did then. F
              • While alone Intel is stable, the combination of Intel chip + Intel motherboard is virtually unbreakable. It's so solid, I think that combination can rival the best of Apple's hardware.

                Problem being, what if 64bit really takes off in a few years. Then you'd rather settle for an Opteron now.

                • In a few years it will be time to replace this server anyway. For now it only runs samba and Postgres. While it may be nice to have more than 4 gigs of address space per process for Postgres, I do not think it is vital.
                  Now for the Asterix server we are thinking about it is a much harder question. For that box stability is so important but at the same time recording calls in speex and tracking them could take a lot of CPU power and RAM. For that box I want triple redundant PS and UPS's and RAID 0+1 with SATA
          • The K5 is a risc processor with an x86 front end on it. The last chip that was AMD's version of an intel chip was the AMD 486. The k6 and k7 only have accessible registers and an instruction set in common with intel processors (WRT k6, AMD's implementation of socket 7 is close but not identical, although it is backwards compatible.) If you buy a motherboard with a via chipset you don't have to talk to via unless you run reference drivers, you only need contact the motherboard manufacturer. I admit that inte
          • Only too right. I got a chipset from AMD at first (Athlon), before they handed it over to the chipset manufacturers. It was pretty stable and had all the right features. Only then came VIA with the chipset (which messed up the USB support, as everybody now knows). Now with the Athlon 64 they did not sell their own chipset, but it took quite some time for the chipset and motherboard manufacturers to get it right. And even now they are still trying to get up to par with the latest Pentium offerings (RAID, Gig
          • Well, in a way, you _do_ get the most important part of the motherboard directly on the Athlon 64.

            And historically, the biggest reason for performance differences and various other issues on motherboards, has been the memory controller. That's for example what used to make Intel's chipsets rock, and Via's suck, or why the NForce 2 quickly became _the_ choice for Athlon XP chipsets.

            Now that AMD has moved that on the CPU itself, you'll notice that in all benchmarks all motherboards perform the same. (1-2% d
      • Intel was "over" RAMBUS for a couple years now, or at least, they haven't been trying to be RAMBUS exclusive for at least that long.
    • Plus, the fact that the benchmarks don't show anything too exciting doesn't help either.

      Let's also not forget VIA's rather dismal history with regards to compatibility, reliability and stability.

  • Yay new chipsets! (Score:1, Interesting)

    by Anonymous Coward
    Just what I always wanted, totally new chipsets with totally new incompatibilites, bugs, and other weirdnesses. What do you want to bet that it takes them one or two firmware revisions to make them work with some major video card vendor's product or another, or that they'll only boot with the pentium chips out now, and you'll have to somehow borrow a processor to flash an updated bios with any new processor IDs that come out say the week after you got the motherboard?
    • This post wasn't interesting, it was insightfull - its true, VIA always has problems with their chipsets until a few BIOS upgrades/driver fixes down the road. Oooh...I just can't wait for the 4in1's that break systems over it's knees. I don't really think this is going to really get people like myself (the budget gamer) interested. Frankly, I don't play that many games and those that I do load up now and again play perfectly fine on my AMD based system.

      However, the market for these products is going to b
      • Re:Yay new chipsets! (Score:5, Informative)

        by Billly Gates ( 198444 ) on Monday January 31, 2005 @03:21PM (#11530957) Journal
        The infamous bug with nvidia was due to nvidias cards not being AGP 2.0 standards compliant in terms of watt usage. Via blamed Nvidia and Nvidia blamed VIA. Intel owned 95% of the market then and Nvidia only tested the geforce with Intel boards. People assumed VIA was just unstable compared to Intel as a result and some still believe it today.

        Vendors like Dell and IBM stuck with Intel as a result.

        Also there was a scam 4 years ago when the athlon boards including defective capacitators that would explode. Most cheaper motherboard makers prefered VIA/AMD solutions due to the cheaper price, also picked the bad capitators. Consumers assumed it was VIAs fault stuck with Intel. A few them made it into Intel boards too including IBM's desktop line but the press was not big to pick that up.

        Finally in 2005 many business users are seeing through the BS of the early days and VIA is fine.

        VIA is not that bad anymore and nvidia works fine with their boards now.
        • Re:Yay new chipsets! (Score:5, Interesting)

          by Jerf ( 17166 ) on Monday January 31, 2005 @03:49PM (#11531230) Journal
          The infamous bug

          "The" infamous bug? Believe me, it takes more than one infamous bug to gain a well-deserved reputation as a purveyor of crap.

          I hit the one with the Soundblaster 128 + Via motherboard. Turns out Via's PCI was shit, and while the Soundblaster 128 did happen to really whale on the particular way in which it was shit, you could randomly lock the system up (completely unstoppably) with any high PCI load.

          Mind you, this isn't their first PCI chipset. PCI had been out for years; we were just starting to see computers coming out that had no ISA connectors at all.

          That motherboard, as I recall, also had memory issues.

          Almost everything I've owned has been Via because I've been a poor college student or worse, and almost everything I've owned has been crap, except the Asus based computer I have. I bought this cheap laptop with a damned Via chipset, and it is the only laptop (even in the cheap-ass class) that runs so hot it burns you when it is idling. Yup, it's the chipset. I wish to high heaven I could replace it.

          "Not Via", after extensive experience, is now my #1 criterion when buying new computers. I don't even care if they've improved; they screwed up so many times over such a long period of time in such stupid ways that it has to be systematic; unless they restructured if they've been "good lately" it's either luck, or simply that you haven't heard of the errors their stuff has yet.
          • Strange, I've had an SB PCI 128 in a Via KT266A and in my current board, a KT400 and haven't experienced the problems you posted.
            • I had it with the KT133A chipset (Abit's KT7A Mobo).

              USB would occasionally lock solid, and games would occassionally lock solid too, the only time I could rely on it was when I was surfing for pr0n.

              Via went onto my 'do not buy' list, but Im thinking of getting their mini-itx stuff in the future.

              Im with an nforce2 now, no probs whatsoever.
  • There was a slashdot article about Intel buying licensing from Nvidia so they could produce SLI motherboards. Looks like VIA isnt waiting around.

    I thought this year might be void of some good hardware, but a dual core Intel board with sli might be be real in Q205. Nice.
  • by qwertphobia ( 825473 ) on Monday January 31, 2005 @02:36PM (#11530535)
    With the cross-licensing agreements with Intel, will VIA be prohibited from transitioning these technologies into their chipsets for the AMD platform?
  • Can i expect the system to not power on if I have a SCSI card and a PS/2 mouse hooked up like in the last system I had with a VIA chipset? Or have they at long last started making products that work?
    • I second that. Had some trouble with some older VIA AMD chipsets and since then never bought VIA again. In particular on the board for my first Athlon the Realtime Clock would jump back and forth under load (under linux that is).
      There was a patch that workarounded it but that one broke other things (like NFS support, USB support and other stuff that depends on timing).
      I'll stay away from VIA for my linux boxes unless I come across a board that has been timetested long enough under linux to be trusted.
    • I've had problems with a number of VIA chipset based systems. Enough so that unless I had substantial proof that there were major benefits to be had in using them and that I had people I trusted telling me the systems built around them were as stable as the Intel platform I'd steer clear of VIA.
  • by EconolineCrush ( 659729 ) on Monday January 31, 2005 @02:39PM (#11530563)
    The Tech Report has a more thorough review [techreport.com] of the chipset, complete with independent benchmarks.
  • Market? (Score:5, Interesting)

    by Dragoon412 ( 648209 ) on Monday January 31, 2005 @02:41PM (#11530579)
    I wonder - and I'm not being facetious or sarcastic - is there really much of a market left for this sort of a thing?

    Intel has, sadly, been having its own ass handed to it in the high-performance/gaming segment for a year or more, now. No gaming enthusiast with the slightest bit of hardware knowledge, which is apparently the PT's target market, owns a P4 system these days.

    Unless this PT chipset is designed to cut costs for resellers like Dell and Gateway with their high-end machines (and I use the term loosely), I don't see it having any impact at all.
    • I'm not sure that there is much of a market. I am a gaming enthusiast that owns a P4 system. Part of the reason I went with the P4 instead of AMD is because the features I were looking for were only offered from VIA boards. After being burned a few years ago by an AGP bus issue with my ASUS K7V Athlon classic 750Mhz using a VIA chipset, I have vowed to avoid VIA if at all possible. After lots of headaches with that system and gaming, I decided the stability I had always gotten from Intel systems in the
    • Re:Market? (Score:3, Interesting)

      by ninjagin ( 631183 )
      It's a fair question. I think there is a market for it, though.

      That Intel has been having a rough go in high-perf/gaming is pretty obvious, however, as a gaming enthusiast with some hardware knowledge I can say that I'm still running a 2.7 GHz (533) P4 without problems (and I don't OC). Haven't come across anything I could not play.

      I only build a new machine every couple years, but I can say that while many of my pals have built with the newer 800Mhz bus CPUs, they don't really seem to be getting that m

      • PCI express is NOT PCI-x. It is usually abbreviated PCIe. PCI-X is an entirely different standard, that is actually mostly backward compatible with standard PCI, physically and electrically. PCIe boards will not work in PCI slots, nor PCI boards work in PCIe slots, but PCI-X and PCI will generally work both ways.

        I probably would not consider buy VIA anyway. For one, their bus mastering in the past has been substandard or non-existent. A lot of people got burned with VIA and other bits of hardware that
        • Sorry about the confusion. I meant PCI-express.

          My point was supposed to be about caution wrt accepting certain chipset cofigurations against P4 compatibility and compliance with DDR2 for the purpose of gaming.

          My nomenclature got in the way, I guess.

          I was also trying to introduce the experience of the half-dozen gamers and sytem builders in my group.

          I am not an EE, though I do read spectrum. I, too, have had interesting and unusual problems with VIA chipsets in the past, particularly wrt firmware updates
        • Yeah I had big problems mastering my bus on the VIA P3 era chipsets.
          Caused me no end of trouble with DMA CD burners. Nothing like having a drive randomly kick back to a slow PIO mode on boot to cause coasters.
    • Just thought I should chime in - I'm a gamer with a Pentium 4. The reason? Stability. I've had problems with VIA in the past as well. I had bad luck with the AGP support on their P3 boards. I had bad luck with their software drivers. The same board also fried on me (the only mobo I've owned that has). And it was an Abit board, so not a super low end mobo.
      Now I can't comment on the current generation stuff because I haven't used it. I just didn't feel like dealing with the possibility of unstable hardware.
      So
  • Heat reduction? (Score:1, Interesting)

    by stanleypane ( 729903 )
    I read another article about the PT's linked on the homepage of the El Reg website.. I don't remember the article mentioning much about it's advantages/disadvantages compared to the P4 or Athlon line of processors.

    I do recall the VIA chipsets running at much cooler temperatures than it's competitors. Perhaps they are on to something. With everyone jumping onto the SFF bandwagon, their chips could prove to be quite befitting in that area. I myself wouldn't mind seeing a 2 GHz processor that doesn't need a
    • I guess it would have helped had I paid attention to the word Chipsets. heh.

      Disregard the ignorance of my last comment. Mod down to 0, off topic and stupid.
      • Oh great, someone already modded me up to a 2, Interesting. Guess I'm not the only one that glazed over that article and mixed it up. heh. For about an hour or so, I was thinking they released a new line of processors. ugh.

        MOD PARENT DOWN ALREADY! Hurry, I don't want to risk embarassment on Slashdot. My reputation all over will be ruined!
    • Check out this review [hothardware.com] to learn a little bit more about the Pentium M. According to their information, and what I saw on Intel's website - the Pentium M's can have a juntion temperature of 100 degrees C (where the proc meets the paste). They ran their Pentium M at 35.5 degrees C overclocked to 2.53GHz!!! They said that P4s idle at 53 degrees C. So I would say this is pretty good. Someone else is more than welcome to find other temperatures for a more balanced comparison.
  • Doesn't anyone at VIA stop and think for a second before deciding on some name for a new product? I own a Taurus PT911 semi-automatic pistol [gunrush.com] for personal protection and in that particular case the name is kind of cool because I get to joke about pitying the fool who dares attack me, not having to call 911 and stuff, but to name a chipset PT is just lame. Chrysler is no better with their Chrysler PT Cruiser [chrysler.com]. Imagine some guy asking a girl: "Hey, wanna go for a ride in my pity cruiser?" What were they thinkin

  • by Anonymous Coward on Monday January 31, 2005 @03:12PM (#11530853)
    Here are some more sites covering it... all about the same content, really:
    viperlair.com [viperlair.com]
    hardocp.com [hardocp.com]
    techreport.com [techreport.com]
    thetechzone.com [thetechzone.com]
    tweaktown.com [tweaktown.com]
    thetechzone.com [thetechzone.com]
    hothardware.com [hothardware.com]
    hexus.net [hexus.net]
    pcper.com [pcper.com]
    legionhardware.com [legionhardware.com]
    thetechlounge.com [thetechlounge.com]
    bigbruin.com [bigbruin.com]
  • I have no idea what IGP is, and I consider myself to be fairly well-informed as far as parts and new technologies go (que up all the elitist and fanboi jokes).

    Honestly, I DID do a search on IGP: you can find it here: http://www.google.com/search?&q=define%3Aigp [google.com]. I think I speak for a lot of us when I say, "Please stop assuming everyone in the world knows what the latest acronym du'jour stands for!"

    -theGreater.

    PS: After doing some addtional subtractive googles, it means "Integrated Graphics ()Pr
  • VIA couldn't even get PCI working right.

    Now you can have PCI lockups and DMA errors many times faster with VIA PCI Express!
  • Yukky!! Eww! Every experience I've had with VIA chipsets has been pretty bad... interrupt sharing especially...
    • Actually, VIA's AMD chipsets have gotten pretty good, especially their AMD64 chipsets, since AMD's integrated memory controller removes one of the trickiest portions of chipset design from VIA's hands. My ASUS K8V Deluxe has performed very well under 64-bit Linux. That said, given a choice, I'll generally spend a few dollars more for a nVidia chipset board, just because nVidia has been such a good Linux supporter (64-bit video drivers).

      If forced to buy Intel's grossly inefficient P4 blast furnace, though
  • Better colours (Score:1, Interesting)

    by Anonymous Coward
  • Better bus handling? (Score:4, Informative)

    by rusty0101 ( 565565 ) on Monday January 31, 2005 @03:47PM (#11531201) Homepage Journal
    The last few Via chipset motherboards I have had, have had deadlock issues with the IDE bus. Capturing live video at mpeg2 speeds would cause a random lockup that required a hard reset to resolve.

    When trying to figure out why, I ended up trying a third party (DFI) IDE board to see if that would resolve the issue. It did not, which suggests that the problem is actually with something at the motherboard on the Via Chipset. I ultimately decided to move to NForce2 boards for my video work.

    I would hop that these issues have been addressed with the new Via chipsets, but I think it would be worthwhile to run some extended testing before you can't return any board with the chipset on it.

    -Rusty
    • This is a common problem with the AGP driver. I tore my hair out for a week before I found out how to fix it. Switching the VIA AGP driver to the standard Microsoft one stopped all the crashes/freezes/reboots.
  • It didn't mention anything in the review - will it support RAID on multiple drives off a single port multiplier? Where are these port multipliers likely to be found, affixed to the case's drive cage?
  • Via reminds me (Score:3, Interesting)

    by MemoryDragon ( 544441 ) on Monday January 31, 2005 @07:48PM (#11534687)
    on the dreaded KT133A chipset which could never be stabilized, after burning through two such boards, and constantly having locks and IDE problems, I went for the a much cheaper SiS based board and suddenly that was the first Athlon board I ever owned which ran totally stable. (and still does after almost four years and 3 processor upgrades)
    Via is a no buy criterion for me everytime I see something from Via I try too look for other options. Last time that was, was a few weeks ago, when I ditched my long term plans of waiting for Via to bring out a decent C3 combo and went for a Mac Mini purchase for my silent server needs.
    • Same here. KT133 (non A) forced me to look for something else as it was way too problematic (very unstable, BSODs, crackling audio, file copy errors, lockups, USB only half working, ...) They were so unusable that there should have been a class action lawsuit, or refunds or replacements for something else. That was the most grossly defective computer part I ever bought (not talking about DOA but just "not working as it should"), and nothing was ever done about it. I've used a lot of SiS boards too, and the
      • Actually it can come worse.... I worked at a small software company back then, and the boss, bought a bunch of computers which had a combo of Via KT133A boards and Deskstar harddisks... The minute I saw this I warned him and told him to replace that part. He refused due to monetary reasons... The troubles this machines gave us were in the long run 10 times more expensive than buying the replacement parts and putting them into the machines... None of the machines worked correctly, harddisks started to fail a
  • I didn't think I was distracted, but when I read the the first sentence of this posting as "The VatIcAn is announcing their new PT series of chip sets at the masses."

    I thought for a moment that the Pope might be giving some competition to the unholy trinity of Gates, Barret, and Dell.

    Oh well, maybe after some white smoke clears...
  • There's better coverage on Anandtech [anandtech.com].

Arithmetic is being able to count up to twenty without taking off your shoes. -- Mickey Mouse

Working...