Software vs Hardware
|
Prev: First Next-Gen CELL Processor: 2 PPEs - 32 SPEs - at least 1 Teraflop
Next: old microcode listings
From: Robert Myers on 13 Dec 2006 13:34 Dennis M. O'Connor wrote: > > These are gamers. It's a hobby. That means there is no > "reasonableness" criteria to apply: just whether it is fun. > -- Gaming is not the only market where there are no "reasonableness" criteria to apply. It's only a question of priorities. Robert.
From: Thor Lancelot Simon on 13 Dec 2006 14:29 In article <1166018014.741002(a)nnrp1.phx1.gblx.net>, Dennis M. O'Connor <dmoc(a)globalcrossing.net> wrote: >"Terje Mathisen" <terje.mathisen(a)hda.hydro.com> wrote ... >> Robert Myers wrote: >>> This whole discussion is surprising. Implementations of the TCP/IP >>> stack in hardware can be bought off the shelf. Every contributor to >>> this thread must know that. There's a card out especially for gamers >>> that implements a network stack. >>> >>> If there were any edge to anything else, somebody'd be doing it, if >>> only for the gamers. If it isn't being done, it's because there's no >>> advantage to it. >> >> Robert, that 'Killer NIC' is very close to pure bullshit, i.e. they sel it >> because they can make money on gullible gamers. > >Actually, HardOCP did some blind testing with and without it, >and on some games (WoW, for example) it did give an >improvement in gameplay. The article is on at Well, of course it does -- the card has QoS profiles tuned for specific games, and, even worse, the benchmark compares to an Ethernet adapter that lacks features (or isn't using them) present on $40-$60 modern cards from Intel, Marvell, and others, such as checksum and segmentation offload and interrupt mitigation. Of course, what the benchmarkers _don't_ bother to do is construct a QoS profile for the application using Windows' built-in QoS facility. I see little reason to think that, if they did so, the performance of the "Killer NIC" would really be different at all. -- Thor Lancelot Simon tls(a)rek.tjls.com "The liberties...lose much of their value whenever those who have greater private means are permitted to use their advantages to control the course of public debate." -John Rawls
From: sylvek on 13 Dec 2006 19:02 "Dennis M. O'Connor" <dmoc(a)globalcrossing.net> wrote: > Actually, HardOCP did some blind testing with and without it, HardOCP with their kooky "blind" testing is going to become an equivalent of "Sterophile" from the sound reproduction field. The next step in reducing latency will be done by putting "Shakti stones" on the twisted pair cable. Check out http://www.shakti-innovations.com/ , SHAKTI On-Lines are known to improve sound quality, image quality and horsepower of the internal combustion engines. Small conceptual step to improving latency of the Ethernet. The step after next will be putting Mpingo disks from http://www.shunmook.com/ somewhere on the network board. I always had faith in the American salespeople. ;-) S
From: Dennis M. O'Connor on 13 Dec 2006 19:23 "sylvek" <__THINK__(a)hotmail.com> wrote in message news:qw0gh.50653$HO4.14659(a)newsfe06.phx... > "Dennis M. O'Connor" <dmoc(a)globalcrossing.net> wrote: >> Actually, HardOCP did some blind testing with and without it, > > HardOCP with their kooky "blind" testing is going to become > an equivalent of "Sterophile" from the sound reproduction field. Wow, there's a well reasoned critique of their article ... NOT. -- Dennis M. O'Connor
From: Tim McCaffrey on 14 Dec 2006 13:18
In article <1165876041.056672.101830(a)16g2000cwy.googlegroups.com>, rohit.nadig(a)gmail.com says... > >Hello Comrades, > >I propose the following Premise: >It seems to me that much of the features that would benefit the average >user of a computer could be better implemented in hardware than in >software. Currently most of these features (multimedia, >networking/communication) are implemented as software applications. > CISC vs. RISC all over again. Inboard or Outboard processing all over again. (Boy, I've been around that wheel a few times). As for the remark about the cache: See Terje's sig. It is relatively easy to design and test a fast integer unit, same for cache. Do you know how hard it is to get a TCP/IP implementation that interoperates with the rest of the world? There is no standard spec, just a bunch of RFCs. If you follow all of them, the result probably will not work, much less interoperate. BTW, a few of those TCP/IP experts Nick mentioned work down the hall from me, so don't give up hope Nick. Anyway, by the time you get your Network accelerator, or video accelerator to work, the standards have changed, the fast integer/cache implementation has gone through two tech turns, and you cost 10x as much. The technology needs to reach a point of diminishing returns before "squeezing blood from a stone" approaches like this become viable. - Tim NOT speaking for Unisys. |