{ "numMessagesInTopic": 5, "nextInTime": 1499, "senderId": "Dwn-l_1oi1h5sicr9P5EzL6ndcvSWebjoR6oaweGVOME4O2DE8iEXL9_NkJGrzuXw5sy-yy07JlqfslkAeWL2Ow9WZjpiENYlKg", "systemMessage": false, "subject": "Re: Problems with converter boxes", "from": ""libert001" <starpicker@...>", "authorName": "libert001", "msgSnippet": "Why do you try not the 61.76 drivers,they are also special for the T221,i work on a FX4000 with no other. ", "msgId": 1498, "profile": "libert001", "topicId": 1496, "spamInfo": { "reason": "12", "isSpam": false }, "replyTo": "LIST", "userId": 174735083, "messageBody": "
>60)
\n> Hi all,
\n>
\n> I spent a very frustrating evening yesterday trying to get 3-stripe
\n> 48Hz to work, and with a genuine IBM converter box to boot! Things
\n> that I found, in no particular order:
\n>
\n> 1) My homemade converter box was almost as good as the genuine IBM
\n> converter box in terms of the highest stable pixel clock reached.
\n> Homemade = 300 Mhz, IBM = 305 - 315Mhz.
\n>
\n> 2) The genuine IBM box, with the original IBM converter box (LFH-
\n
> to monitor (LFH-60) cable, still could not reach the magic 328Mhzmode
\n> pixel clock rate needed for 48Hz 3-stripe mode. This is rather
\n> disturbing. I am suspecting a) the dual-link cable I am using -- it
\n> is a expensive one from Gefen, 1-foot long to minimize length, but
\n> has no ferrite inductor blocks on it, or b) the 6800GTo card itself
\n> may not be able to do much over 300Mhz pixel clock in dual link
\n
> even though 180-190Mhz in single link is routinely achieved.a
\n>
\n> 3) The Nvidia driver has a whole bunch of issues. I am still using
\n
> 77.xx or so, definintely not 80.xx Detonator driver b/c I foundthat
\n
> the 80.xx drivers did not let me span a DG3 in 2-head mode.windows,
\n>
\n> As I mentioned before, I got 3-stripe mode working (sort of), at
\n> 42Hz, but only in DualView mode (ie. 2 separate monitors in
\n
> 1 at 2624x2400 and 1 at 1216x2400). When I replaced the home-madestripe
\n> converter box with the genuine one I got off eBay, and booted into
\n> Windows, the 42Hz 3-stripe Dualview mode was still ok, as expected.
\n>
\n> However after messing around with the driver settings in an attempt
\n> to get 48Hz 3-stripe to work, I could not get back to the stable 3-
\n> stripe 42Hz mode!
\n>
\n> 4) I tried to get dual-dual link to work after I gave up on 3-
\n
> 48Hz at least for the evening. Setting the EDID to 5, I hooked upa
\n> each converter box to a different 6800GTo card (I have 2 in my
\n> machine). This is obviously different from Wilson who I think used
\n
> Quadro with 2 dual-links on one card.both
\n> This did not work well at all -- I got 1 half of the monitor
\n> displaying the expected 1920 x 2400 @ 48Hz, but the right half of
\n> the monitor was blank/black. Occasionally there would be a flash of
\n> something, but I can't be sure. This is disappointing because the
\n> pixel clock needed for 1920x2400 @ 48Hz is easily achievable by
\n
> converter boxes, it may again be a driver issue.T221
\n>
\n> I am sort of at a loss now, and was debating getting one of the new
\n> 7900GT gaming cards which have dual-dual link. But from the recent
\n> discussion stating that new Nvidia cards have problems with the
\n
> (probably from the drivers) I'm wary of even that!
\n>
\n> Again -- if there is anyone here who has gotten 3-stripe mode to
\n> work with a non-Quadro dual-link card please let me know.
\n>
\n> Also another question for Wilson and Flup: When you did get 48Hz
\n> working, what kind of dual-link cable were you using from the card
\n> to the converter box? This may be the weak link in my system. Does
\n> IBM usually include a dual-link cable for that portion?
\n>
\n> Yeang
\n>