Author: Chris Whittington
Date: 02:33:58 10/28/97
Go up one level in this thread
On October 27, 1997 at 22:50:48, Keith Ian Price wrote: >On October 27, 1997 at 08:49:35, Chris Whittington wrote: > >>>Well, I had planned to, but I can't get it to work with Auto232. I know >>>what the problem is, but I am not pleased with what I have to do to get >>>it to work. The problem is that although it's a DOS program, it requires >>>a mouse. You cannot work it through the keyboard, and in fact if there >>>is no mouse, it just quits. >> >>No it doesn't, it gives you a message about the lack of mouse and quits. > >Yes, this is what I meant. This is OK. Not a complaint. > >>>Well, I have two computers for doing >>>autoplay, both are Pentium 150s. The one I tried to run it on has a >>>modem on com2, so I tried to run it in /c232 mode, hoping that it >>>wouldn't need a mouse, simply to respond as Black. It just locked up and >>>I had to power cycle. >> >>The presence of the auto232 software is enough to cause a crash rather >>than a nice messge if you run without a mouse. I think this is to do >>with the auto232 drivers using a port normally given to the mouse, and >>this confuses the DOS call to 'do we have a mouse' - the results can be >>unpredictable. I don't intend doing anything about this, since normal >>users get a nice message about their mouse, and power users of auto232, >>basically can cope :) Connect your mouse. > >I can cope, and I already surmised this. Hope your technical support >people can cope, if CSTal is successful. And before I go on; >congratulations on the good results in the first three rounds. If this >continues, Thorsten will be totally jazzed when he gets back from Paris! >No one will be able to keep him quiet for weeks. Good job. > > >>> Well in the other, I have removed the modem, and I >>>thought I'd just buy a converter for the auto232 cable and run it off >>>com2, but in that one I have an ATI video card, and the mouse works as >>>an eraser as it moves around the screen, so I suppose I can get it >>>working, but I am tired of messing with it right now. >> >>My word, you are having problems ! >> >>Sorry, but CSTal was designed from the start to be mouse driven. >>Personally, I find the UI's of the disk and manual chess programs to be >>just dreadfull; I wanted to do something different. > >No problem with mouse driven. Not an issue, although keyboard shortcuts >would be nice. Its got keyboard shortcuts: alt-M force move alt-Q quit alt-L load game alt-D change 2d/3d - loads of others f1, f2, f3, f4, f5 change screen layouts even typing your move in algebraic works (although you don't see the input), this is how the auto232 works. there's load more have you tried clicking all over the screen with the mouse ? You'll be surprised at how much stuff there is ..... Story: when I got a fine JApanese ar about four years ago, I discovered they had an engineering concept - latent design - these are not things they tell you about, but things you find out about, slowly, as you use it. you don;t get overloaded with gallons of stuff, but you keep finding new things you can do. I liked this idea. CSTal is designed the same way. > >>So your first problem is contention between your modem and auto232 >>interface, requiring both the available com ports, and leaving no >>hardware gap for the mouse. > >I can remove the modem. I don't use it. I have to get an adapter to run >off com2, though, since Chrilly's cable only supports D9 connectors. > >>Your second problem is lack of an appropriate VESA driver (CSTal is in >>SVGA). This you can only solve by asking your machine manufacturor. > >I had surmised that also, and tried the /vesa switch to see if it made a >difference, but it didn't. I think I can get a vesa driver from ATI, but >this is going to be a big problem for you, since it affects non-"power" >users as well. Its a big problem for all SVGA programs in DOS, and getting worse. its become very common for manufacturors to boot their customers direct into windows, and not even bother to set up the dos environment. these pc's, especially when used by naive users will fall over. And naive user always says: "but all my other programs work" We use the Watcom compiler library functions to handle the screen. Our vesa operation first looks for the vesa driver software; if it can't find that it has a go at running without it (possible with some cards). Another recent problem is with dual page mode svga cards. As ever the poor user gets left in dual page mode when he's in DOS. Not many dos programs know about dual page mode. Solution ? Find the driver software (different for each card type) that flips the thing back into single page mode. Anyway, tech support here knwos more about this than me ..... You can email for answers. > >>Your third problem (the saitek board) I'm going to have to ask Thorsten >>to answer next week after Paris. I can't test it here because our Saitek >>board is kaput now. > >I would guess it works with the Saitek board. I don't have a Saitek >board. I have a Smartboard II from TASC, and AFAIK it's one of the >"known auto-sensory boards". At least CG5, and Chessica (of course) and >CM5.5k think so, with built-in support. CSTal also, according to your >ad, but I can wait for Thorsten. Ah, ok; it should work with this one also. Again I've seen it. > >>However the code is in there to run this board, and I've seen it work. I >>do recollect, either buried in the on-line manual or somewhere, a >>warning that the Saitek code used the parallel printer port and could >>fail under some circumstances - I don't remember what, but persumably >>some contention on this port. > >Nothing else on the port, and nothing using IRQ 7. > >>>I guess Chris is >>>too busy with Paris right now to get back about the other problems, but >>>I'd say that it doesn't much matter if everyone else is running on >>>60000Mhz Crays, if his program is so full of bugs, >> >>Hey, hey, hey. I will not be so foolish to say that it has no bugs, but >>we've spent a great deal of effort on QA, I can assure you. > >Yes, I apologize for the tone here, but I was frustrated at the time, >even though I had figured out what was necessary. Imagine those who >can't figure it out. Its maddening, I know. Which is why with all our other products (Bridge, Go, Shogi) we are doing win95 and no other pc support. The chess is working up on win95 also. At least win95 is moderately successful in the plug and play department. > >>Further, apart from the Saitek issue, which is not clear at the moment. >>Your 'riddled with bugs' appears to be (a) you disconnecting your mouse, >>and (b) a hardware/ VESA driver issue. Both of these are outside my >>control. And are not bugs. > >As a professional software bug remover for my company, and a past and >future (possibly) beta tester for OS/2 for IBM, I disagree. If the >software still stated that the mouse port was in use, and quit, then no >bug. Black screen of death=bug. OK, but theer is a limit to what we can trap, no ? I'll admit to not being too bothered about power users with no mouse (when they know the program wants a mouse), who then plug another device into the mouse port which has the effect of confusing the mouse interrogation software. In fact the first thing we do is say hello to the mouse. The mouse (if present) then returns its coordinates, always the center of the screen. Double the values, and magic, we have the screen size. useful for svga, because the user could have been in any one of many screen sizes. Nice bug some time back. Microsoft mouse driver 9.1 returned the upper left coordinates (0,0). Double those and you have the screen size ...... Microsoft withdrew 9.1 fairly quickly. Now is it a bug that we didn't test for (0,0) ? I guess so, but there must be a limit to what we can guess is going to go wrong, no ? > Eraser mouse=bug; message stating that >the video card does not support Vesa standard codes and quit=No bug. As >for being out of your control, I don't think so. If you mean that as a >chessplayer, and not a programmer, you do not know how to check such >things safely, then perhaps you need to consult with some PC programming >experts, as many other programs can make these checks. I'm the chess programmer, and I write the chess code. We employ programmer 'experts' to do the UI stuff. I'm not aware of foolproof ways to test vesa capability. In fact I don't think there are any, let me know if you know different ... And judging by the number of times 'top selling' stuff (entertainment software) fails on some of our kit here, I don't think anybody else does either. > >> >>We will be putting CSTal Paris up for free download to customers on our >>www site, the day that Paris ends. > >This was one reason I was posting. I was hoping to get the patches in >the download version, to save your tech support people (is that you?) >some trouble. No, its Steve Hart. > >>The Paris version contains a few changes to the general program usage, >>but, probably more importantly, contains some search improvements that >>were created after the release of the original version. These >>improvements appear to be fairly substantial. > >Judging from the first three rounds, I don't think anyone will argue >with you. You are doing wonders for TC's soul. Thanks, but ... TC's soul is shot to hell :) Chris Whittington >
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.