sichermn@beach.csulb.edu (Jeff Sicherman) (11/27/90)
In a previous posting, someone mentioned a file with information on setting up a dual monitor based system. I believe it was ftp-able. I tried to email a request for it but unfortunately my mailer chokes on some types of addresses. I'd rather not consume bandwidth with what appears to be a common thread, so good someone send me the file or direct me to its location ? Thanks Jeff Sicherman sichermn@beach.csulb.edu
sichermn@beach.csulb.edu (Jeff Sicherman) (11/27/90)
I'm sure this has been discussed to death but I missed it all so ... Want to use a dual monitor system. Have mono monitor (true blue) with Herc adaptor (true ?). Want to add an EGA monitor and card (yes, I said and mean EGA, this is a compatibility issue for a target system so dont argue/flame about superiority of VGA). Are there any special requirements on EGA card (if its the type that emulates/supports herc doesnt that interfere with the herc itself ?). Are any adaptors excluded. Can any adaptors support both monitors concurrently ? Jeff Sicherman sichermn@beach.csulb.edu
krab@iesd.auc.dk (Kresten Krab Thorup) (11/29/90)
Check your dealer to be sure... I suppose most `modern' EGA/VGA adaptors are capalable of being used with a Hercules Monochrome adaptor. Kresten
jdb@reef.cis.ufl.edu (Brian K. W. Hook) (02/07/91)
Someone recently posted that if I have a 16-bit VGA card and an 8-bit monochrome card in my system that my VGA will automatically be throttled down to 8-bit...this would become an unacceptable bottleneck in my IO (I have a 386-33). If so, is there a way around it? Brian
nee@cf_su14.Sbi.Com (Robert Nee) (02/08/91)
> Someone recently posted that if I have a 16-bit VGA card and an 8-bit > monochrome card in my system that my VGA will automatically be throttled > down to 8-bit...this would become an unacceptable bottleneck in my > IO (I have a 386-33). If so, is there a way around it? As far as I know there isn't. The solution would be a monochrome display adapter with a 16-bit interface. This would allow both displays to coexist and operate at full 16-bit speed. But alas monochrome is dead... Who would want a 16bit mono board anyway? I WOULD! Robert F. Nee <nee@cf_su20.Sbi.Com>
jim@rwsys.lonestar.org (James Wyatt KA5VJL) (02/09/91)
In article <26762@uflorida.cis.ufl.EDU> jdb@reef.cis.ufl.edu (Brian K. W. Hook) writes: > >Someone recently posted that if I have a 16-bit VGA card and an 8-bit >monochrome card in my system that my VGA will automatically be throttled >down to 8-bit...this would become an unacceptable bottleneck in my >IO (I have a 386-33). If so, is there a way around it? We use SuperVGA boards for running Windows and add a modified MDA board for running CodeView (ToadView!) to debug. An unmodified MDA really slows video access even when you aren't debugging, so we added a switch on the back. If we had Turbo switches on the front I'd use that and a relay/gate. The switch disables memory and IO reads to the MDA board, so the BIOS won't detect it during the POST. Just flip the switch and reset (NOT Ctl-Alt-Del on most machines) to choose speed or debugging. To add the enable switch, you'll need: 1ea, DPDT or DPST (rare) switch (small!) (Radio Shack 275-626) 2ea, 10K Ohm 1/4W resistors (2/5 of Radio Shack 271-1339) 2ft, 30ga wire-wrap wire (Radio Shack 278-503 or -501) 1. Hardware work: Remove the mounting bracket from the board. Drill a hole near the bottom for the switch (make sure it clears anything near the edge of the board) and install the switch in the bracket. Reattach the mounting bracket to the MDA board. 2. Cut the traces going to the B12 and B14 contacts on the 62-pin bus connector. Wire the switch and two resistors like so: cut trace-\ v B12 >-----------\ X /---------*----> to -SMEMR uses | | | 10K O O O \--/\/\/---*---> +5V [add switch]-> | O O O /--\/\/\---/ | | | 10K B14 >------------/ X \--------*----> to -IOR uses ^ cut trace-/ 3. Install the board and test for proper operation. The resistors are to pull-up the memory and IO read lines if %sMOS is used while not loading the bus lines too much. Hope this helps someone else as much as much as it has us. - jim ---- James Wyatt (KA5VJL) - Standard disclaimer applies... (H)817-595-0571 {letni.lonestar.org,merch.tandy.com}!rwsys.lonestar.org!jim (W)817-390-2864
jdb@reef.cis.ufl.edu (Brian K. W. Hook) (02/10/91)
In article <166@cf_su20.cf_su10.Sbi.COM> nee@cf_su14.Sbi.Com (Robert Nee) writes: >> Someone recently posted that if I have a 16-bit VGA card and an 8-bit >> monochrome card in my system that my VGA will automatically be throttled >> down to 8-bit...this would become an unacceptable bottleneck in my >> IO (I have a 386-33). If so, is there a way around it? > >As far as I know there isn't. The solution would be a monochrome >display adapter with a 16-bit interface. This would allow both >displays to coexist and operate at full 16-bit speed. But alas >monochrome is dead... Who would want a 16bit mono board anyway? I just discovered from someone over at comp.windows.ms.programmer that supposedly the VGA cards out there all operate at 8-bit bandwidth when IN GRAPHICS MODE....is this true? Even the 16-bit ones do. Brian
nee@cf_su14.Sbi.Com (Robert Nee) (02/11/91)
> I just discovered from someone over at comp.windows.ms.programmer that > supposedly the VGA cards out there all operate at 8-bit bandwidth > when IN GRAPHICS MODE....is this true? Even the 16-bit ones do. No. This is not true. However all VGA cards will operate in 8-bit mode ALL THE TIME if they are installed in a system with an 8-bit MDA board. This puts about a 32% drain in display performance on my system. I just spent a day searching around a computer fair for an MDA with a 16-bit interface. No luck though. I think I may give up. Robert F. Nee <nee@cf_su20.Sbi.Com>
douglass@davidsys.com (02/11/91)
In article <26810@uflorida.cis.ufl.EDU>, jdb@reef.cis.ufl.edu (Brian K. W. Hook) writes: > In article <166@cf_su20.cf_su10.Sbi.COM> nee@cf_su14.Sbi.Com (Robert Nee) writes: >>> Someone recently posted that if I have a 16-bit VGA card and an 8-bit >>> monochrome card in my system that my VGA will automatically be throttled >>> down to 8-bit...this would become an unacceptable bottleneck in my >>> IO (I have a 386-33). If so, is there a way around it? >>As far as I know there isn't. The solution would be a monochrome >>display adapter with a 16-bit interface. This would allow both >>displays to coexist and operate at full 16-bit speed. But alas >>monochrome is dead... Who would want a 16bit mono board anyway? > I just discovered from someone over at comp.windows.ms.programmer that > supposedly the VGA cards out there all operate at 8-bit bandwidth > when IN GRAPHICS MODE....is this true? Even the 16-bit ones do. I'm afraid that the only ways I know of to get to the graphics memory involve (steps are pseudo-operations): 1) selecting the bit-plane to write to, 2) setting up 8-bit color mask registers on E/VGA card, 3) writing pixels to that color plane, 4) go back to step 1 until all color planes have been updated. Note step 2. I believe that [EV]GA has a hardware limitation in 16-color graphics modes which cripples normal VGA graphics to 8-bit operations. The 256-color modes may not have this problem, I don't know. (Any corrections gladly accepted). Also, I have done some testing of some systems I have available and found no difference in performance to my 16-bit VgaWonder with an IBM monochrome adapter in or out. Testing indicates (on *my* system, YMMV): 16-bit VGA fastest (only text mode tested, of course) 8-bit M[DG]A (or CGA) takes twice as long to write to as 16-bit VGA IBM EGA (8-bit) three times as long as 16-bit VGA These results gathered on 6 and 8 MHz 80286 AT clones, 16 MHz Compaq 386, 16 and 33 MHz 80386 clones. Test consisted of an assembly program that moved screenfuls of data from main memory to the screen 2000 times and counted timer ticks. Sorry to run on so long, but I just can't understand why an 8-bit card should slow down a 16-bit card!?! At least it doesn't in my case. > > Brian -- -{JD}- Jeff (douglass@davidsys.com) /* My opinions are my own. Who else would want them? */ "Never count on the inevitable until it happens. . ." "So therefore a pointer to dev/nul (the nul device) is a NULL pointer?"
phil@cs.mcgill.ca (Philip LOCONG) (02/12/91)
In article <8130@davidsys.com> douglass@davidsys.com writes: > >Also, I have done some testing of some systems I have available and >found no difference in performance to my 16-bit VgaWonder with an >IBM monochrome adapter in or out. >Testing indicates (on *my* system, YMMV): >16-bit VGA fastest (only text mode tested, of course) >8-bit M[DG]A (or CGA) takes twice as long to write to as 16-bit VGA >IBM EGA (8-bit) three times as long as 16-bit VGA > >These results gathered on 6 and 8 MHz 80286 AT clones, 16 MHz Compaq 386, >16 and 33 MHz 80386 clones. >Test consisted of an assembly program that moved screenfuls of data >from main memory to the screen 2000 times and counted timer ticks. > >Sorry to run on so long, but I just can't understand why an 8-bit >card should slow down a 16-bit card!?! At least it doesn't in my case. > As earlier posts suggested it, the ISA bus specifications force the bus to run each 128k section of RAM entirely in 8-bit mode or entirely in 16-bit mode, that means the A-B section has to be either 8 or 16-bit. E[V]GA uses A000-B000 and B800-C000 for RAM while the MDA uses B000-B100. This means if you have both, the RAM will interface with the bus in 8-bit mode. And then, there is also the fact that many 16-bit VGA cards (not all) have an internal 8-bit data path which becomes the actual bottleneck wether you use 8 or 16-bit mode. On the other hand, an original IBM monochrome adapter uses 4k of static RAM and has a data rate of 1.8 M bytes/sec. This will give the VGA cards of today a much better performance than MDA even when run at 8-bit. Then there's also the BIOS in the C-D section... Philippe Locong phil@bart.cs.mcgill.ca
nee@cf_su14.Salomon.Com (Robert Nee) (02/12/91)
>Sorry to run on so long, but I just can't understand why an 8-bit >card should slow down a 16-bit card!?! At least it doesn't in my case. Perhaps your VGA card isn't running in 16-bit mode anyway. I have seen some cases where the VGA card wouldn't kick into 16-bit mode no matter how few cards were installed in the system. Try this. Run your VGA card's diagnostics with the MDA card in and out. Most will tell you if the card is running in 16 or 8 bit mode. If it is running in 8 bit mode, certain operations WILL be slower. Tests on my system for graphics intensive applications show about a 32% dip in performance with an MDA present. Robert F. Nee <nee@cf_su20.Sbi.Com>
jdb@reef.cis.ufl.edu (Brian K. W. Hook) (02/13/91)
It seems that we never reached a conclusion whether or not a 16-bit VGA card will be throttled down to 8-bit by having an MGA card plugged in also, thus, if you reply, please answer the following questions: 1. How fast does the 16-bit VGA transfer information normally? 2. If a 16-bit VGA is throttled down to 8-bit, would a 16-bit MGA help things? 3. If it would, how hard would it be for EE friend of mine to construct a 16-bit MGA? Brian
rodman@sgi.com (Paul K. Rodman) (02/13/91)
Hello Netters, Sorry to pester the net with this, but I have an old Compaq 286 "portable" (about the size of a small satchel) with a problem. It has a 20Mb hard disk that makes a terrible rattling racket. There seems to be no problem with the disk otherwise. I recall someone telling me that there is in fact, a simple fix for this problem if you have access to the appropriate tools to get at some pesky loose spring clip or something. I've reached the point where this noise is intolerable, and I'm willing to risk losing the disk so any info on how this problem is fixed would be wonderfully received... Thanks, Mail responses to: Paul Rodman Silicon Graphics, Inc. rodman@sgi.com 415 335 7209 -- Paul K. Rodman Advanced Systems Division Silicon Graphics, Inc.
bmarsh@cod.NOSC.MIL (William C. Marsh) (02/14/91)
In article <169@cf_su20.cf_su10.Sbi.COM> nee@cf_su14.Sbi.Com (Robert Nee) writes: >> I just discovered from someone over at comp.windows.ms.programmer that >> supposedly the VGA cards out there all operate at 8-bit bandwidth >> when IN GRAPHICS MODE....is this true? Even the 16-bit ones do. > >No. This is not true. However all VGA cards will operate in 8-bit mode >ALL THE TIME if they are installed in a system with an 8-bit MDA board. Sorry, the EGA/VGA interface is only 8 bits wide, so even if your VGA connects to the '16 bit' connector, the board itself can only do 8 bits at a time. If you look at the memory architecture on the board, you will also notice that it is (256, 512, ot 1024) x 8 bits. All the cards 'fake' 16 bit accesses in graphics mode. And anyway, all memory accesses to video memory is sync'd to H or V retrace time, so 16 vr. 8 bits really don't amount to much. (Even in the VRAM adapters too! What a waste!) Bill -- Bill Marsh, Naval Ocean Systems Center, San Diego, CA {arpa,mil}net: bmarsh@cod.nosc.mil uucp: {ihnp4,akgua,decvax,dcdwest,ucbvax}!sdcsvax!nosc!bmarsh "If you are not part of the solution, you're part of the problem..."
nee@cf_su14.Salomon.Com (Robert Nee) (02/14/91)
> And anyway, all memory accesses to video memory is sync'd to H or V retrace > time, so 16 vr. 8 bits really don't amount to much. Hmmm. On my system it amounts to about a 32% speed degredation. This is also very noticable in windows. I am not just spouting useless figures. Do you have a monochome adapter? Have you run you own benchmarks? Do you want a copy of the ones I have been using? What are the results in your own tests? What video board are you using? Mr. Marsh's theories are sound, but something falls short in pactice. Robert F. Nee <nee@cf_su20.Sbi.Com>
douglass@davidsys.com (02/14/91)
In article <1991Feb12.010248.7563@cs.mcgill.ca>, phil@cs.mcgill.ca (Philip LOCONG) writes: > In article <8130@davidsys.com> douglass@davidsys.com writes: >> >>Also, I have done some testing of some systems I have available and >>found no difference in performance to my 16-bit VgaWonder with an >>IBM monochrome adapter in or out. >>Testing indicates (on *my* systems, YMMV): >>16-bit VGA fastest (only text mode tested, of course) >>8-bit M[DG]A (or CGA) takes twice as long to write to as 16-bit VGA >>IBM EGA (8-bit) three times as long as 16-bit VGA >> >>These results gathered on 6 and 8 MHz 80286 AT clones, 16 MHz Compaq 386, >>16 and 33 MHz 80386 clones. >>Sorry to run on so long, but I just can't understand why an 8-bit >>card should slow down a 16-bit card!?! At least it doesn't in my case. > >As earlier posts suggested it, the ISA bus specifications force the bus >to run each 128k section of RAM entirely in 8-bit mode or entirely in >16-bit mode, that means the A-B section has to be either 8 or 16-bit. Excuse me, but you're *WRONG*. I've already run a test that indicates otherwise. I know of projects where this 'limitation' was 'overcome'. The 'problem' lies in the timing of the address lines coming from the 16-bit slot. To put it simply, if the card responds quickly enough, there is NO PROBLEM. (If the motherboard and/or card is not designed well enough, the card can never respond 'quickly enough'.) In other words: Numbers talk (see above tests). I would be interested to know of other cards that support fast access. (or the converse). >[ stuff proving that VGA and monochrome live within 128K deleted ] >run at 8-bit. Then there's also the BIOS in the C-D section... Now, with all of that out of the way, we all realize that the Video ROM for the [EV]GA (at C000) is 8-bit. Right? By your argument, that would mean that all of C000-DFFF must be 8-bit? Including your 16-bit ExPANded memory board (EMS)? Including your 16-bit (memory_mapped) Network board (WD, 3COM, etc)? Including your 16-bit nifty fast (memory_mapped) SCSI board? Including your 16-bit (memory_mapped) tape backup board? Try doing benchmarks, people. It's really not that diffucult. > >Philippe Locong >phil@bart.cs.mcgill.ca -- -{JD}- Jeff (douglass@davidsys.com) David Systems, Sunnyvale CA,(408)720-8000 /* My opinions are my own. Who else would want them? */ "Never count on the inevitable until it happens. . ." "So therefore a pointer to dev/nul (the nul device) is a NULL pointer?"
mlord@bwdls58.bnr.ca (Mark Lord) (02/15/91)
In article <2819@cod.NOSC.MIL> bmarsh@cod.nosc.mil.UUCP (William C. Marsh) writes: <Sorry, the EGA/VGA interface is only 8 bits wide, so even if your VGA <connects to the '16 bit' connector, the board itself can only do 8 bits <at a time. If you look at the memory architecture on the board, you will <also notice that it is (256, 512, ot 1024) x 8 bits. All the cards 'fake' <16 bit accesses in graphics mode. Utter rubbish! The paradise vga card I have has memory that is 16 bits wide, and it does indeed use 16-bit writes to update it (in text mode at the least). <And anyway, all memory accesses to video memory is sync'd to H or V retrace <time, so 16 vr. 8 bits really don't amount to much. (Even in the VRAM <adapters too! What a waste!) I don't know that much about the VRAM things, but the whole idea of using dual-ported VRAMs is to overcome the need to wait for retrace cycles before writing, so this statement is probably as accurate as the first paragraph was. -- ___Mark S. Lord__________________________________________ | ..uunet!bnrgate!mlord%bmerh724 | Climb Free Or Die (NH) | | MLORD@BNR.CA Ottawa, Ontario | Personal views only. | |________________________________|________________________|
bmarsh@cod.NOSC.MIL (William C. Marsh) (02/15/91)
In article <176@cf_su20.cf_su10.Sbi.COM> nee@cf_su14.Salomon.Com (Robert Nee) writes: >> And anyway, all memory accesses to video memory is sync'd to H or V retrace >> time, so 16 vr. 8 bits really don't amount to much. >Hmmm. On my system it amounts to about a 32% speed degredation. This >is also very noticable in windows. I am not just spouting useless >figures. You cannot use a windows based program for performing graphics benchmarks. There is too much system in the way. To properly test graphic board performance, you must talk directly to the hardware. No BIOS, No Windows, No DV, etc. >Do you have a monochome adapter? Have you run you own benchmarks? Yes, and Yes. I write graphics animation software as a side job, and I do not notice any speed degredation with a mono monitor or not. >Do you want a copy of the ones I have been using? What are the >results in your own tests? What video board are you using? I use a Video 7 (now Headland) VRAM VGA with 512K on it, and a no-name clone of the Hercules (Actually, the manual is a copy of the Hercules manual, with the name blanked out. Really 'no-name' ;-). Look into any book about the hardware of the VGA: Everything is 8 bits wide! Bit masks, plane latches, etc. I would like to see your tests. Bill -- Bill Marsh, Naval Ocean Systems Center, San Diego, CA {arpa,mil}net: bmarsh@cod.nosc.mil uucp: {ihnp4,akgua,decvax,dcdwest,ucbvax}!sdcsvax!nosc!bmarsh "If you are not part of the solution, you're part of the problem..."
nee@cf_su14.Salomon.Com (Robert Nee) (02/15/91)
> Try doing benchmarks, people. It's really not that diffucult. Grunt... Wheez... Puff... Puff... Whew! That was tough... ;-) There, I did the benchmarks. With my MDA board in and with it out. I get 32% better performance in graphics benchmarks with it out. Am I *WRONG* also? > To put it simply, if the card responds quickly enough, there is NO PROBLEM. Isn't that dandy. Which card? The 8-bit or the 16-bit? My MDA board cost me $8. My VGA board cost $300. Do I need a new, better VGA to make my $8 MDA work, or vica versa? In addition, to anyone whose benchmarks show no performance change. Are you sure that your board is running in 16-bit mode to begin with? My VGA board's Diags report 16-bit mode only after I set the proper switches and ONLY if there is no MDA board in my system. My VGA (V7 VRAM) CLEARLY works better if it can get into 16-bit RAM mode. With MY MDA it can't. Does anyone know of an MDA that will let it? Thanks. Robert F. Nee <nee@cf_su20.Sbi.Com>
andyross@infopls.chi.il.us (Andrew Rossmann) (02/18/91)
jdb@reef.cis.ufl.edu (Brian K. W. Hook) writes: > It seems that we never reached a conclusion whether or not a 16-bit VGA > card will be throttled down to 8-bit by having an MGA card plugged in > also, thus, if you reply, please answer the following questions: > > 1. How fast does the 16-bit VGA transfer information normally? > 2. If a 16-bit VGA is throttled down to 8-bit, would a 16-bit MGA > help things? > 3. If it would, how hard would it be for EE friend of mine to > construct a 16-bit MGA? Here is some text taken from 'Prgrammers Guide to the EGA and VGA cards - Second Edition' by Richard F. Ferraro. (pages 614-615). 13.2.6 16-bit Memory Transers Designing a 16-bit interface for memroy transfers is not trivial on the AT bus. The PC bus was designed by IBM. Once it was carved in stone there was no turning back. There is a problem with the latched and unlatched address bus on the AT. These two address buses are passed to the adapters through the AT connectors. When 16-bit memory accesses are desired, a tri-stated line called MEM16 must be pulled low by the adapter, indicating to the host processor that a 16-bit transfer is desired. The host has to do some fancy footwork to determine whether it should output two 8-bit bytes on data lines 0-7 or one 16-bit word on data lines 0-15. In order to have enough time, the AT bus requires that the MEM16 line be asserted shortly after the address lines settle. The address lines are used by the adaptor to determine whether it should respond at all to the bus address. Thus, it has to be decoded by the adaptor. Consider the following example. The VGA display memory resides at address A0000-B000 hex in the PC memory address space. The host wants to read a word from the VGA at address A1000. It exerts the address A1000 on the bus and waits for a bit. During this time, all devices on the memory bus have to look at the address, A1000, to determine if they should respond. In this case, the VGA address decoder would detect that A1000 resides between A000 and AFFFF and so gears itself up. It still does not know if the host wants to read or write. SInce this VGA is a 16-bit device, it wants to exert MEM16 to alert the host to this fact. As soon as it decoes the A0000 hex, it sends out the low signal on the MEM16 line. The host in turn identifies the MEM16line and sends the memory read pulse. Then it waits for a bit before it reads the 16-bit data on the data bus bits 0-15. During this second wait, this VGA has to use the read pule to enable the output of its memory (previously addressed at A1000) and send the 16-bit signal onto the data bus bits 0-15. The problem is that decoding the address takes time for the VGA, and it has to send out the MEM16 line quickly. The address cannot be decoded until the address lines settle. On the PC, there are two sets of address lines on the PC being latched and unlatched. The unlatched lines settle before MEM16 has to be asserted, while the latched lines settle after MEM16 has to be asserted. We can rule out using the latched address lines. The problem is that no complete set of unlatched address lines are available, the lowest being address bit 17. As a result, the address decoding circuitry cannot detect memory locations that are smaller than 128Kbytes. Therefore, the VGA can only detect that a memory access is required somewhere between A0000 and BFFFF before it can decide whether to respond with a MEM16 signal. Unfortunately, in dual monitor systems, a Hercules card may well be residing at B0000 with unpredictable results. I have yet to see a 16-bit Hercules implementation. The MEM16 line is another reason why 16-bit VGAs will operate in an 8-bit mode, while in AT systems with bus speeds greater than 8MHz, they will not operate in a 16-bit mode. The increased speed of the bus and the subsequent decreased time for the MEM16 line to be asserted exceeds the decoding rate of the VGA. End of quote. In reality, some VGA cards will run 16-bit at >8MHz bus speeds. It depends on how well they are designed. --------------- Andrew Rossmann | Sysop of Infoplus BBS, +1 708 537 0247 andyross@infopls.chi.il.us | Infoplus Support, latest version available uunet!ddsw1!infopls!andyross | by logging in as infoplus.
bmarsh@cod.NOSC.MIL (William C. Marsh) (02/21/91)
In article <5621@bwdls58.UUCP> mlord@bwdls58.bnr.ca (Mark Lord) writes: >In article <2819@cod.NOSC.MIL> bmarsh@cod.nosc.mil.UUCP (William C. Marsh) writes: ><Sorry, the EGA/VGA interface is only 8 bits wide, so even if your VGA ><connects to the '16 bit' connector, the board itself can only do 8 bits ><at a time. If you look at the memory architecture on the board, you will ><also notice that it is (256, 512, ot 1024) x 8 bits. All the cards 'fake' ><16 bit accesses in graphics mode. >Utter rubbish! The paradise vga card I have has memory that is 16 bits wide, >and it does indeed use 16-bit writes to update it (in text mode at the least). That doesn't mean the VGA actually has 16 bit wide data paths to/from display memory. Read *any* EGA/VGA reference and you can see that the internal data path is 8 bits. ><And anyway, all memory accesses to video memory is sync'd to H or V retrace ><time, so 16 vr. 8 bits really don't amount to much. (Even in the VRAM ><adapters too! What a waste!) >I don't know that much about the VRAM things, but the whole idea of using >dual-ported VRAMs is to overcome the need to wait for retrace cycles before >writing, so this statement is probably as accurate as the first paragraph was. A 10 Mhz, No Wait 286 can move approxamatly 2 Mbyte/sec with memory move instructions, and yet, the 'fastest' VGA card claims to have about 11K chars/sec. These tests do not use the BIOS interface, but go directly to the memory buffer. How can you explain two orders of magnitude? The basic fact is that the video boards on the PC are *not* a true multi-port ram, just a simple arbitrator, with the retrace getting the highest priority. (I think I remember reading in my VRAM Tech Ref. that you get one cycle out of five during active display times). Bill -- Bill Marsh, Naval Ocean Systems Center, San Diego, CA {arpa,mil}net: bmarsh@cod.nosc.mil uucp: {ihnp4,akgua,decvax,dcdwest,ucbvax}!sdcsvax!nosc!bmarsh "If you are not part of the solution, you're part of the problem..."
phil@cs.mcgill.ca (Philip LOCONG) (02/21/91)
> [endless discussion about 8 vs 16 internal data path and poor > performance caused by retrace and uselessness (! english is NOT my first > language...) of VRAM etc...] This is all very interesting but I see many "all" and "no VGA card has...". I think the "all 8-bit" argument might have been true in the past but now, many cards are now explicitly advertised as having 16-bit (even 32!) internal data paths, some even claim to have a video memory cache (check a logix-2000L add). I think recent VGA cards might be very decent. Philippe Locong phil@bart.cs.McGill.CA
phil@brahms.amd.com (Phil Ngai) (02/21/91)
In article <8345@davidsys.com> douglass@davidsys.com writes: |Excuse me, but you're *WRONG*. |I've already run a test that indicates otherwise. |I know of projects where this 'limitation' was 'overcome'. |The 'problem' lies in the timing of the address lines coming from the |16-bit slot. |To put it simply, if the card responds quickly enough, there is NO PROBLEM. |(If the motherboard and/or card is not designed well enough, the card can |never respond 'quickly enough'.) That's the whole point. Some motherboards are implemented in such a way that the 128 Kbyte limit exists. If you want to talk about selected systems, sure, you can get around it. But in general, it is not safe to mix 8 and 16-bit devices within a block of 128 Kbytes. |Now, with all of that out of the way, we all realize that the Video ROM |for the [EV]GA (at C000) is 8-bit. Right? |By your argument, that would mean that all of C000-DFFF must be 8-bit? |Including your 16-bit ExPANded memory board (EMS)? |Including your 16-bit (memory_mapped) Network board (WD, 3COM, etc)? |Including your 16-bit nifty fast (memory_mapped) SCSI board? |Including your 16-bit (memory_mapped) tape backup board? Indeed, it must be. And I find that the EMS boards I look at have an option to put them into an 8-bit mode. Other products may have this controlled by their BIOS. My network board is 8-bit also. |Try doing benchmarks, people. It's really not that diffucult. Try looking at some motherboard chip specs. It's much more useful than taking a small sample and assuming it applies to all machines. -- It is time for California farmers to share in the water shortages that the cities have endured for 5 years.
fatooley@cc.usu.edu (Edgar Tooley) (06/25/91)
I have a 386 compatable with a vga card that I would like to attach to a large monitor in a class room setting. The monitor is a 20" monocrome which requires a composite input from the computer. My question is, what options are availiable to do this I would like to have the same display on both monitors with one showing VGA color and the other monocrome. Thanks Edgar Tooley Utah State University Life Span Learning