jdm@hodge.UUCP (jdm) (07/31/89)
Three questions:
1) Could someone explain the difference between the 8250, 16450,
and 16550 UARTs?
2) Is it possible to replace 16450 UARTs in a serial I/O board
with 16550 UARTs with no modifications to the circuit?
3) Give a few mail order references to where one might purchase
16550 UARTs and the typical price of same.
Thanks,
--
"I'm an anthropologist, not a computer systems architect, damit!"
jdm@hodge.cts.com [uunet zardoz vdelta crash]!hodge!jdm
James D. Murray, Ethnounixologist TEL: (714) 998-7750 Ext. 129
Hodge Computer Research Corporation FAX: (714) 921-8038
1588 North Batavia Street
Orange, California 92667 USAchasm@attctc.Dallas.TX.US (Charles Marslett) (08/01/89)
In article <21200@hodge.UUCP>, jdm@hodge.UUCP (jdm) writes: > Three questions: > > 1) Could someone explain the difference between the 8250, 16450, > and 16550 UARTs? They are all 8250s with a few new features. The 16450 adds some additional flags and a register for the programmer to save some local data (if he needs it). The 16550 (I think) adds several bytes of buffering for the recieved data so interrupt latency is not nearly so critical. > 2) Is it possible to replace 16450 UARTs in a serial I/O board > with 16550 UARTs with no modifications to the circuit? I am almost certain that this is true. We have been testing some hardware/ software designs at STB, and I do not believe there are any jumpers or other bits of configuring hardware around the UART sockets (and the card takes both 16450s and 16550s, even mixed). > 3) Give a few mail order references to where one might purchase > 16550 UARTs and the typical price of same. Sorry, don't know of any -- I just play with the stuff [Andi buys it...]. > Thanks, > jdm@hodge.cts.com [uunet zardoz vdelta crash]!hodge!jdm > James D. Murray, Ethnounixologist TEL: (714) 998-7750 Ext. 129 > Hodge Computer Research Corporation FAX: (714) 921-8038 > 1588 North Batavia Street > Orange, California 92667 USA Charles Marslett STB Systems, Inc. <-- apply all standard disclaimers
yhe@zippy.eecs.umich.edu (Youda He) (08/01/89)
Does any one knows how to get a 8250 replacement? My Modem haas a 8250 on it, it will not work at 10M Hz, So I had to slow down to 6 MHz mode in order to ues the modem. I guess 16450/16550 can not directly replace 8250?
stevel@eleazar.dartmouth.edu (Steve Ligett) (08/04/89)
In article <21200@hodge.UUCP> jdm@hodge.UUCP (jdm) writes: >Three questions: > > 1) Could someone explain the difference between the 8250, 16450, > and 16550 UARTs? ... I'm not a UART expert, but I have an application note from National Semiconductor titled "A Comparison of the INS8250, NS16450 and NS16550A Series of UARTs". Application Note 493. The introductory pragraph says: National currently produces seven different versions of the INS8250 UART. Functionally, these poarts appear to be the same, however, there are differences that the designer and purchaser need to understand. For each version, this document provides a brief overview of their distinct characteristics, a detailed function and timing section, a discussion of software compatibility issues and the AC timing prarmeters. Steve Ligett steve.ligett@dartmouth.edu or (decvax harvard linus true)!dartvax!steve.ligett
william.richardson@canremote.uucp (WILLIAM RICHARDSON) (08/10/89)
The 16550A is a teriffic UART, It will replace the 16450 or 82450 and give you a 6K buffer that is supported by Telix. It has allowed my USR HST 14.4 to work properly @ 38400bps without errors. --- ~ EZ-Reader 1.20 ~ * QNet 1.03a3: CanConfMail: . HQBBS / ~ Toronto ~ Canada ~ 416y566y1765
Ralf.Brown@B.GP.CS.CMU.EDU (08/10/89)
In article <89080907301955@masnet.uucp>, william.richardson@canremote.uucp (WILLIAM RICHARDSON) wrote: }The 16550A is a teriffic UART, It will replace the 16450 or 82450 and }give you a 6K buffer that is supported by Telix. It's a 16-byte buffer, not 6144 bytes.... But even that makes a huge difference (38400 would generate interrupts at the same rate as 2400 without the buffer). -- UUCP: {ucbvax,harvard}!cs.cmu.edu!ralf -=-=-=-=- Voice: (412) 268-3053 (school) ARPA: ralf@cs.cmu.edu BIT: ralf%cs.cmu.edu@CMUCCVMA FIDO: Ralf Brown 1:129/46 FAX: available on request Disclaimer? I claimed something? "Drama is life with the dull bits left out." -- Alfred Hitchcock
karl@ddsw1.MCS.COM (Karl Denninger) (08/11/89)
>Item 4910 (0 resps) by william.richardson at canremote.uucp on Thu 10 Aug 89 >13:18 >[WILLIAM RICHARDSON] Subject: Re: 16450/16550 UART question > >The 16550A is a teriffic UART, It will replace the 16450 or 82450 and >give you a 6K buffer that is supported by Telix. It has allowed my USR >HST 14.4 to work properly @ 38400bps without errors. 16550As have 16 character FIFOs, not 6KB! They require software that knows how to work with them -- other than that, they are teriffic. -- Karl Denninger (karl@ddsw1.MCS.COM, <well-connected>!ddsw1!karl) Public Access Data Line: [+1 312 566-8911], Voice: [+1 312 566-8910] Macro Computer Solutions, Inc. "Quality Solutions at a Fair Price"
palowoda@fiver.UUCP (Bob Palowoda) (08/14/89)
In article <89081106353440@masnet.uucp>, william.richardson@canremote.uucp (WILLIAM RICHARDSON) writes: > The 16550A is a teriffic UART, It will replace the 16450 or 82450 and > give you a 6K buffer that is supported by Telix. It has allowed my USR ^^^ This is a 16 byte FIFO. > HST 14.4 to work properly @ 38400bps without errors. ^^^^^^^ If this is a V32 modem I doubt that this is the real bps effective out the modem. Most likely it's the throughput of the modem measured with compression done on the data. Just think if your sending compressed data to the modem already. If you really want to spec a modem modem fairly state: cps/sec uncompressed data cps/sec compressed data. ---Bob > --- > ~ EZ-Reader 1.20 ~ > > * QNet 1.03a3: CanConfMail: . HQBBS / ~ Toronto ~ Canada ~ 416y566y1765 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ This is kind of interesting. Did this come of a PC-Board type bbs? -- Bob Palowoda *Home of Fiver BBS* login: bbs Home {sun,dasiy}!ys2!fiver!palowoda (415)-623-8809 1200/2400 Work {sun,pyramid,decwrl}!megatest!palowoda (415)-623-8806 1200/2400/9600/19200 Voice: (415)-623-7495 Public access UNIX system
william.richardson@canremote.uucp (WILLIAM RICHARDSON) (08/14/89)
So I've found out. Now I can't remember where I heard it was 6K, either way; it works. Colin S. mentioned he only uses 8 charaters in Telix 3.1x Still if it works it works & a hole lot cheaper than a 386 machine! --- ~ EZ-Reader 1.20 ~ What a nag I must be..... * QNet 1.03a3: CanConfMail: . HQBBS / ~ Toronto ~ Canada ~ 416y566y1765
thaler@shorty.CS.WISC.EDU (Maurice Thaler) (08/16/89)
In article <123@fiver.UUCP> palowoda@fiver.UUCP (Bob Palowoda) writes: >In article <89081106353440@masnet.uucp>, william.richardson@canremote.uucp (WILLIAM RICHARDSON) writes: >> The 16550A is a teriffic UART, It will replace the 16450 or 82450 and >> give you a 6K buffer that is supported by Telix. It has allowed my USR > ^^^ This is a 16 byte FIFO. > >> HST 14.4 to work properly @ 38400bps without errors. > ^^^^^^^ > If this is a V32 modem I doubt that this is the real > bps effective out the modem. Most likely it's the > throughput of the modem measured with compression done > on the data. Just think if your sending compressed data > to the modem already. If you really want to spec a modem > modem fairly state: > cps/sec uncompressed data > cps/sec compressed data. > Actually, most of the high speed modem manufacturers recommend that you set your high speed modem to lock its speed to the terminal at a rate at least DOUBLE the actual connect rate. So if you are connected at 9600, you set your terminal to 19200. If you can achieve higher than 9600 (the latest HST modems can do something like 14000, than 38400 is the baud rate you would lock your terminal too. As was stated, with MNP protocol, you get an increase in speed through data compression, but this does not happen unless you have both terminals locked to higher rates than the actual modem rate. It has been my experience with two HST modems locked to 19200 that I can get about 1120 bps with ZMODEM. I understand with the new HST's this can go up to about 1400bps- actual throughput. It certainly is easy to confuse people with this kind of arrangement. One important thing to keep in mind if you attempt this higher baud rate locking is YOU MUST IMPLEMENT HARDWARE HANDSHAKING! If you don't, flow control falls apart completely. On the HST you do this by entering AT&H1 Maurice Thaler SYSOP Audio Projects BBS (608) 836-9473 SYSOP Power Board BBS (608) 222-8842
wek@point.UUCP (Bill Kuykendall) (08/16/89)
>> The 16550A is a teriffic UART, It will replace the 16450 or 82450 and >> give you a 6K buffer that is supported by Telix. It has allowed my USR > ^^^ This is a 16 byte FIFO. > >> HST 14.4 to work properly @ 38400bps without errors. > ^^^^^^^ > If this is a V32 modem I doubt that this is the real > bps effective out the modem. Most likely it's the > throughput of the modem measured with compression done > on the data. Just think if your sending compressed data > to the modem already. If you really want to spec a modem > modem fairly state: > cps/sec uncompressed data > cps/sec compressed data. > Good point, but he wasn't spec'ing the modem, he was spec'ing the uart. In fairness, neither had 38.4k throughput though. The HST 14.4 has an effective throughput of about 26k with max compression. Your mileage will vary. The HST is not V.32 standard. USRobotics makes a V.32, and a Dual Standard with V.32 *and* HST 14.4, but word has it that they're not ready for prime time yet. Bill Kuykendall ...ddsw1!point!wek