Jump to content
HWBOT Community Forums

Antinomy

Crew
 View Profile  See their activity

  • Posts

    1985

  • Joined

  • Last visited

  • Days Won

    53

 Content Type 

Posts posted by Antinomy

    PCmark05 limits

    in Benchmark software

    Posted · Edited by Antinomy

    so what about if it were 220.157mbs

    or has it got to be 220.000

    The rule is "220 is the limit".

     

    This means - if you get 220, you are very lucky. Otherwise you should look for not exceeding this value which means - below 220. Can't get exact - get it lower.

     

    Otherwise we will always argue on how much can "not much" be ;)

    As for me, I know that this difference is much lower than even 1 PCMark point, so it doesn't make sense.

     

    But dura lex sed lex.

     

    P.S. I think that the limit issue should be written on the benchmark page. In the rules section. As a rule - all rules should be mentioned on the appropriate pages, making it easier for everyone.

    Strange 9800GT

    in General hardware discussion

    Posted

    Yep, already've seen this ;) the first daty I got this card. And in my system GPU-Z showed the same - 0 ROPs and shader. Don't know waht ZFess has done to make it tell the truth :D Everest was only capable of showing that SLI is disabled and not even a word on the chip - seems it was in deep shock :)

    Videocard specifications requests : Geforce 6100/6150 specs and category are wrong

    in Support

    Posted

    that doesn't matter imho. the card has a different id and the specs (pci-e line) are slightly different.

    so there's a seperat category for it.

    it doesn't play a role if the performance is similar.

     

    if we merge them people will ask for a new category because the 6150se isn't there anmyore. ;)

    Well, I don't think this makes sense. We are talking about integrated video. So there's no card. There is a chipset and a GPU part. You say the ID's different. So, if we take a two-chip chipset, for example, the old 6100 and different secondary south bridges - 405, 410, 430. They have different ID's but the primary south is the same - 6100. We put the in one category despite the south bridge. So why does this change when we go to single-chip south bridge?

    Take a look in the driver:

    NVIDIA_DEV.02E0.01 = "NVIDIA GeForce 7600 GT"

    NVIDIA_DEV.02E1.01 = "NVIDIA GeForce 7600 GS"

    NVIDIA_DEV.02E2.01 = "NVIDIA GeForce 7300 GT"

    NVIDIA_DEV.02E3.01 = "NVIDIA GeForce 7900 GS "

    NVIDIA_DEV.02E4.01 = "NVIDIA GeForce 7950 GT "

    NVIDIA_DEV.0390.01 = "NVIDIA GeForce 7650 GS"

    NVIDIA_DEV.0391.01 = "NVIDIA GeForce 7600 GT "

    NVIDIA_DEV.0392.01 = "NVIDIA GeForce 7600 GS "

    NVIDIA_DEV.0393.01 = "NVIDIA GeForce 7300 GT "

    NVIDIA_DEV.0394.01 = "NVIDIA GeForce 7600 LE"

    NVIDIA_DEV.0395.01 = "NVIDIA GeForce 7300 GT "

     

    You can see two 7600GT, 7600GS, 7300GT, so why don't we make two categories, cause

    the card has a different id. it doesn't play a role if the performance is similar.
    In fact, the performance is not "similar", it's the same! Though, the performance and chip of the 6100 when C51 and 61 are being compared is not.

    As for the specs - should we separate the ATI All-in-wonder series in a separate category - their specs are different, like the ViVo support, aren't they? The answer is - it doesn't give anything in 3D performance. Like the extra PCI-E 1x doesn't.

     

    Well, I could show a lot more same cards with different ID's (like 9500GT), but I think the example was enough to show my opinion. The same platform (I think the AMD and Intel chipsets may be separated), the same videochip (but not in 6100 - there are two if them) - different categories.

     

    You can name it like 6150SE/6100 (C61 only) category. :) or a full list of supported chipsets, I mentioned above. The same is for 7050 if they have same video. I think there should be an addition to what is meaned as a different card. And there should be both cases - discrete and integrated viewed separately. IMHO, the GTS250 and 9800GTX+ have less reasons to be separated than 6100 C51 and C61 ;)

    Videocard specifications requests : Geforce 6100/6150 specs and category are wrong

    in Support

    Posted

    huh, sounds a bit complicated to me. :P

    i've updated/edited a few things.

    Yep, thank you. That's better, but still not perfect. Why did you leave the 6150 having changed everything else? It's C51. And delete the memory frequency of the 6150LE and 6150SE - cause it's integrated.

     

    The thing I'm trying to show is that the C61 6100 are much closer to 6150SE rather than C51 6100. So the category separation is wrong. Look at the 3DMark 06 (GPU intensive load) results - the top5 is on C61. Didn't look the rest.

     

    I'll try to get a board on C51 6100 to check the speed difference.

    Videocard specifications requests : Geforce 6100/6150 specs and category are wrong

    in Support

    Posted · Edited by Antinomy
    update

    Ticket ID: 606

     

    Priority: Medium

     

    I\'ve been benching these chipsets and found out that there is a number of 6100/6150 chipsets. That's what NVIDIA driver says us:

    NVIDIA_DEV.0240.01 = \"NVIDIA GeForce 6150\"

    NVIDIA_DEV.0241.01 = \"NVIDIA GeForce 6150 LE\"\

    NVIDIA_DEV.0242.01 = \"NVIDIA GeForce 6100\"

    NVIDIA_DEV.03D0.01 = \"NVIDIA GeForce 6150SE nForce 430\"

    NVIDIA_DEV.03D1.01 = \"NVIDIA GeForce 6100 nForce 405\"

    NVIDIA_DEV.03D2.01 = \"NVIDIA GeForce 6100 nForce 400\"

    NVIDIA_DEV.03D5.01 = \"NVIDIA GeForce 6100 nForce 420\"

     

    Whe can see two series of chipsets - the ones with DevID 02xx and with 03xx. The difference is that the first family is dual chip (two south bridges) and the second are single chip. But also the first series is based on Crush51 and the second on Crush61. The new Crush as I read (very poor info on them) has some Z-buffer optimizations enabled. And it's speed increased on same clock. But all of the second series chipsets have their VGA clock at 425MHz. The difference between the 6100 405, 430 and 400 is only the functionality of the second south bridge like USB et cetera. The difference between gen.2 6100 and 6150SE is that 6100 supports 17 PCI-E line (16+1) and 6150SE supports 18 (16+1+1). All these things make no influence on the subject - the graphics. They use the same platform and all specs are similar.

     

    So I would like to ask to combine all Crush 61 results in one category because the difference between 6100 Crush51 and 6100 Crish51 is much more than between 6100 C61 and 6150 SE.

     

    As for the first generation, the 6150 has a different clock, 475 which is worth a separate category. And as for the 6150LE - I haven't found any info on it yet about it's differences.

     

    And the category info needs to be updated, the 6100 are integrated and aren't based on NV43 :)

     

    P.S. as I just found out, the difference between 6100 and 6150 is VC1/H.264 decoding support for 6150. Also it supports newer DXVA (2.0) and DVI stuff and so on. Nothing vitale for 3D benching ;) maybe combine the Crush51 in two categories, the 425 clock and 475?

    Team runs

    in Team Discussions

    Posted

    I've got a question about group benching, but not with team mates. The question on sharing hardware within a team is clear and I agree with the rules. My nearest team mate though is about 7000-7500km from me, others even more.

     

    But last weekend I was having a bench session with another team from my city, with the people I know. Maybe it's the last, maybe it's not, but is hardware sharing between me and another team allowed? We just took the HW out of the bags on the table and oc'ed everything what eyes saw first. I surely understand that they can't share results of one HW within their team and there were a few CPU's of one model, what will be seen by their stepping, when they are out.

     

    But the different team benching is a question, I wanted to figure out. Nobody is pushing his team higher, cause we are in different ones.

    Bwanasoft's insufficient proof in 3DM2001SE

    in Submission & member moderation

    Posted

    but spamming this thread with useless "+1" is not the theme either.
    I agree absolutely.
    If you are going to nitpick the results of somebody else, be prepared to get the favor returned.
    Quite fair. That's why I offered checking out mine.
    he REBENCHED with this CPU and systems to provide better/correct validation
    You mean less suspicious? :)
    but people are linking to scores which no longer exists, showing screens which have already been deleted, blocked OR explained.

    http://hwbot.org/hwbot.post.do?postId=716

    The date is out. Why not just delete Jigit's results, which were very suspicious? Both gave me a tough race. Both suspicious. One a mod, the other banned. Where is the edge? Cause I don't want to ask for whom is it ;)

    then other people join the forum to post "+1" in a thread where they never before visited.
    This thread was created 9th of April. A couple of days ago I never visited this thread, so?
    won't block an Aquamark (or other score) if it has no outlandish result for the hardware used, because the screen is lacking validation
    Mine was though it wasn't outstanding, just №1 in the category. One man didn't know about certain hardware, I didn't have time to make an ORB version.

    Bwanasoft's insufficient proof in 3DM2001SE

    in Submission & member moderation

    Posted

    jmke, you replied on two non-informative posts, but didn't say a word on Gradus' arguments, though asked for them. Isn't that irrational? :)

    Well, I not such a braincracker as Gradus to check every CPU-Z screen, whether it's text boxes are on the right positions, but this is a quite good one.

    Gradus, the AMD Athlon text in the CPU string can be achieved if the MoBo doesn't know the processor. It gets it's CPU string (named in CPU-Z specification) from the frequency table, but if it's o/ced, then it doesn't find a propriate and writes a default text, exactly this one.

    But this can't explain the position of the text.

    And can't explain the screen with "Processo" string. Should be Processor :)

     

    And again, jmke.

    A man shows you an invalid screenshot, but you show him his invalids. In this thread we discuss a certain subject. But you're reversing the situation to the author. Try to find at least 5-10 inproper screens of mine :) that's not the theme.

    Bwanasoft's insufficient proof in 3DM2001SE

    in Submission & member moderation

    Posted

    jmke, Jigit painted screens, the result - ban.

    Bwana painted at least Palomino, the result - deleting them and saying "I won't do like that". Maybe doing in another way, but not getting caught? ;-)

    http://img72.imageshack.us/my.php?image=750pt6.png =-O

    you say that a cut edge is a multimonitor bug? Isn't that too much bugs on one man? :-D

    And you say that the strange white line right under the window is because Bwana has modified a black wallpaper? There's nowhere to get the white there from, except MSpaint, or what he prefers.

     

    I just don't get the clue - if the screens are fake (or what for do they have to be deleted?), then their author is a cheater. If there is a cheater, then he should be banned. Just like Jigit.

     

    We are having a competition between benching teams in our country, and Neoforce at the results said a great phrase, when one result wasn't accepted:

    "You thought you can cheat, and after it didn't work out, you can instead use a correct, but lower result?"

    That's what I wanted to say. One can, others get ban. And no normal explanations this far.

    yes, this is dual monitor setup. yes you can have tasks on other monitor.
    Then why is the paint task on this screen? Or you want to say, that you can select, which tasks should be on the other monitor? :-)

    Bwanasoft's insufficient proof in 3DM2001SE

    in Submission & member moderation

    Posted

    IMHO, no bugs can explain a 0,13um CPUID 681 Palomino, like Gradus said. If I see this screen with the combinations of three parameters. Only after that I will believe.

    Not a screen on my Photoshop projects, but validated by CPU-Z only.

    Yes, there are a number of minor bugs in CPU-Z but this is not a minor bug.

    At least the Palomino case is an approvement. Anyone show me a datasheet were a AMD 0,13um processor with CPUID681 will be called Palomino - I'll shut up. Not a handmade 'course. LOL

×
×
  • Create New...