slerner@sesame.UUCP (Simcha-Yitzchak Lerner) (09/03/85)
**This one's for the line eater**
I am in the process of evaluating _currently_available_ ICE
systems for use with 8088/8086/80286 processors, and have not
had much luck finding a good system.
I would appreciate any experiences/information/rumors you have,
particularly in terms of ease of use and transparency of
different ICE systems.
I looked at two systems a while back. One stole NMI for its
own use, which made it almost unusable for our needs. The other
system disassembled code into the form of JMP/CALL $+offset, and
also only allowed memory addresses to be specified with a
non-segmented address. These two 'features' made this system
much less user friendly than I would care for.
I need a true ICE - a hardware debug card (ala Atron) will not
do.
Excuse my cross-posting this - please respond either via mail or
to net.micro, since I do not subscribe to net.arch or
net.research.
Thanks in advance.
--
Opinions expressed are public domain, and do not belong to Lotus
Development Corp.
----------------------------------------------------------------
Simcha-Yitzchak Lerner
{genrad|ihnp4|ima}!wjh12!talcott!sesame!slerner
{cbosgd|harvard}!talcott!sesame!slerner
slerner%sesame@harvard.ARPA cem@intelca.UUCP (Chuck McManis) (09/06/85)
> **This one's for the line eater** > > I am in the process of evaluating _currently_available_ ICE > systems for use with 8088/8086/80286 processors, and have not > had much luck finding a good system. > > I would appreciate any experiences/information/rumors you have, > particularly in terms of ease of use and transparency of > different ICE systems. If you really want a good ICE unit get an I^2ICE from Intel. It was after all, designed for the part :-) I have used one on my job several times and always found that its operation was identical to that of the "real" chip. --Chuck (* Forget the disclaimer, I work for Intel and I happen to know it is Intel's opinion that all serious debug work on *86 designs should be done with I2ICE [pronounced eye-squared-ice] *) -- - - - D I S C L A I M E R - - - {ihnp4,fortune}!dual\ All opinions expressed herein are my {qantel,idi}-> !intelca!cem own and not those of my employer, my {ucbvax,hao}!hplabs/ friends, or my avocado plant. :-}
phil@amdcad.UUCP (Phil Ngai) (09/07/85)
In article <59@intelca.UUCP> cem@intelca.UUCP (Chuck McManis) writes: >If you really want a good ICE unit get an I^2ICE from Intel. Yes, get an I^2ICE from Intel if you want to spend many bucks ($40,000? For a floppy based system, I think.) and learn a new text editor (what was the name of that brain damaged thing, credit or something) and a whole new operating system (isis, more powerful than a speeding CP/M, or is it the other way around?) Until recently, you had the delightful choice of ASM86, PL/M86, or Pascal. There may be a C now. Maybe. Even so, we thought we wanted one. Then the salesman told us "You're the last customer we'd ship one to. You compete with us." Nice way to do business. I don't think his attitude was condoned by Intel management, but if you are in a business that competes remotely with Intel it might be prudent to consider how vulnerable you want to be. Rumor has it that Intel is coming to their senses and porting their ICE to an IBM-PC host instead of forcing their own weirdness on the world. If I^2ICE is for you, it may be worth waiting for the IBM-PC version. My impression of the whole business is that yes it was very powerful but most of the time overkill for what people really use and need. Depends on what you're doing, of course. I am told by another ICE maker that Intel has this trick of hiding the information you need to make a good ICE for the 80286 except on special "bond out" versions which they don't sell except as part of their ICE. This "bond out" version of the 80286 has extra pads to bring out chip state information necessary to run an ICE. This seems like a dubious technique but Intel seems to be getting away with it. Just my opinion. I speak only for myself. I broke into this machine before AMD implemented their security system and they don't know I wrote this. :-) -- The overseas Chinese are the Jews of Asia. Phil Ngai (408) 749-5720 UUCP: {ucbvax,decwrl,ihnp4,allegra}!amdcad!phil ARPA: amdcad!phil@decwrl.ARPA
brad@gcc-bill.ARPA (Brad Parker) (09/08/85)
In article <3567@amdcad.UUCP> phil@amdcad.UUCP (Phil Ngai) writes: >the last customer we'd ship one to. You compete with us." >Nice way to do business. I've heard at lot about Intel's "attitude problem" - even as far as "having to interview" to buy parts from them. See what working with IBM does to you ;-) >hiding the information you need to make a good ICE for the 80286 >except on special "bond out" versions which they don't sell except >as part of their ICE. This "bond out" version of the 80286 has extra >pads to bring out chip state information necessary to run an ICE. This is true, but you must understand that it can also be looked at as their privaledge to bring out more state for their ICE. It's a neat trick - I've also heard it described as a hack that they had to do because they didn't think before they fab'd. (yuk yuk) We have one of these. To my knowledge, everything Phil says is correct. Besides, he has a great attitude. >Just my opinion. I speak only for myself. I broke into this machine >before AMD implemented their security system and they don't know I >wrote this. :-) Good line! I about fell off my chair. Gosh, you must be one of them there hackers I read about in Time magazine! ;-) wow. I can't believe I did all that editing with a hangover. -- J Bradford Parker uucp: seismo!harvard!gcc-bill!brad "She said you know how to spell AUDACIOUSLY? I could tell I was in love... You want to go to heaven? or would you rather not be saved?" - Lloyd Coal
woof@psivax.UUCP (Hal Schloss) (09/08/85)
In article <3567@amdcad.UUCP> phil@amdcad.UUCP (Phil Ngai) writes: >In article <59@intelca.UUCP> cem@intelca.UUCP (Chuck McManis) writes: >>If you really want a good ICE unit get an I^2ICE from Intel. > >Yes, get an I^2ICE from Intel if you want to spend many bucks ($40,000? >For a floppy based system, I think.) and learn a new text editor (what >was the name of that brain damaged thing, credit or something) and a >whole new operating system (isis, more powerful than a speeding CP/M, >or is it the other way around?) Until recently, you had the delightful >choice of ASM86, PL/M86, or Pascal. There may be a C now. Maybe. > I have had some experience with I^2ICE and believe I can speak with just a little bit of unbiased authority :-) We did some research on the various ICE's available for the 8086-80186-80286 family and it became obvious that the best ICEs one can get are Intel's products. They do have knowledge and special parts available to them, that give Intel a tremendous head start towards a good product. On the other hand they tend to be expensive and don't fit really well into our development environment that centers around our VAX 11/750 running UNIX. If you want they will sell you a lot of equipment to talk to a VMS machine. (feh!) I do program in C and can use the results on the I^2ICE though. I use the C cross compiler package for OASYS in Cambridge, Mass. It would seem that as long as you can get an Intel compatible object file, you can use it on an I^2ICE. In addition you can get (real soon now:-)) I^2ICE for an IBM PC, and something called TRACE86 which looks like a cheaper I^2ICE. If I was buying an I^2ICE again I might consider those two options very strongly. My biggest objection to I^2ICE is the incredible amount of time it seems to take to load code. In conclusion while Intel ICE products are expensive and not incredibly convient to use sometimes, they do the best job of emulating the chip and giving one access to the internal workings of it. -- Hal Schloss (from the Software Lounge at) Pacesetter Systems Inc. {sdcrdcf|ttdica|quad1|scgvaxd|nrcvax|bellcore|logico|rdlvax|ihnp4}!psivax!woof ARPA: ttidca!psivax!woof@rand-unix.arpa
dougp@ISM780.UUCP (09/09/85)
We just acquired an I**2ICE here at Interactive which is XT-hosted. It works. That's just about all you can say about it. The human interface is TERRIBLE, as seems usual with Intel products. There is a kinda prompt line at the bottom of the screen which gives you choices of things to type next in the middle of a command, but I still have all my standard gripes about Intel command parsers (you've got to fully spell keywords like FOREVER, and at times where only one keyword is allowed you still have to type the fool thing, etc.). I don't know if this is a released Intel product or not (we got ours directly from Intel development 'cuz we're doing some 80286 work for them), but they should be out soon. I've heard that the rumored price for the XT/AT interface board is roughly $4K. Yup, you heard me, and this is for a 1/2 sized board! You still need the I**2ICE box and a personality module for the processor you want to debug. These total another $20K. I don't think it works very well on a PC without a hard disk, as all the software and overlays have to be on the same disk and it won't fit on a (360K) floppy. Another alternative is the HP 64000 system with an 80286 emulator. This whole beast (integrated work station w/floppies and the emulator package) costs about $20K and I've heard it's real nice. It is run with programmable function keys whose definitions change depending on where you are in a command. A menu above the keys tells you what's going on. From all I can find out about it, I wish we had one... Cheers, Doug Pintar at Interactive Systems, Inc. (ihnp4!ucla-cs!ism780!dougp) (My employers may agree with what I say, but NEVER publicly...)
jer@peora.UUCP (J. Eric Roskos) (09/10/85)
> I've heard at lot about Intel's "attitude problem" - even as far as > "having to interview" to buy parts from them. See what working with > IBM does to you ;-) If we're going to talk about this subject (I don't know that it has much to do with computer architectures, but where else to talk about it?) there's something else significant about the difference between Intel and Motorola: their method of supplying documentation. Intel has a free service you can subscribe to, by which you get a glossy magazine each month with marketing things in it. Motorola has a paid service (fairly expensive, but worth it) by which you pay a fee and get sent copies of their literature for a year on a subscription basis, as it comes out. I'll take that over the Intel approach anytime. Also, a problem I found when I was in the University environment was that Intel after awhile developed a special program for college faculty & researchers, separate from the one for engineers; instead of the glossy magazine you got items that were supposed to relate to "education". The problem was that at the time I was very actively involved in research on an engineering problem, and didn't WANT a lot of stuff about education; I wanted to know what parts Intel made that could be used in this multiprocessor we were building. I always thought that educational material was kind of insulting, as if people in universities weren't "real" engineers and didn't need real data. I like Motorola's approach: charge a reasonable fee but give you just the FACTS, and let you make the decisions yourself. [They never did answer a few of the crazy questions I asked them about unconventional RAM access, but in general they were very helpful. (Now that I think about it, we used an Intel part for that special RAM... oh well... don't let the details get in the way of the generalizations... :-) )] -- Shyy-Anzr: J. Eric Roskos UUCP: Ofc: ..!{decvax,ucbvax,ihnp4}!vax135!petsd!peora!jer Home: ..!{decvax,ucbvax,ihnp4}!vax135!petsd!peora!jerpc!jer US Mail: MS 795; Perkin-Elmer SDC; 2486 Sand Lake Road, Orlando, FL 32809-7642 "Nalgvzr gbzbeebj, gur cubar'yy evat, naq lbh'yy or ba lbhe jnl. Onpx ubzr va Buvb, gurl jba'g oryvrir lbh..."
david@daisy.UUCP (David Schachter) (10/01/85)
My company, Daisy Systems, is now selling our "Personal Logician AT" bundled with an Intel I2ICE. The package combines our CAE software for hardware design with Intel's ICE expertise for hardware debug. I don't know about prices or that stuff: I'm >just< a programmer. For more info, the address is Daisy Systems, 700 Middlefield Road, Mountain View, CA, 94039-7006. The telephone number is (415) 960- 0123. "I2ICE" and "ICE" are probably trademarks of Intel Corporation. "Personal Logician" and "Personal Logician AT" are probably trademarks of Daisy Systems Corporation. Everything remaining is probably trademarked by somebody. [Why can't the English learn how to speak?]