keithley@applelink.apple.com (Craig Keithley) (11/15/90)
68340 doesn't have a MMU. Dual channel DMA, dual UARTS 68332 has a TPU (Timing Processing Unit - REALLY great for automotive apps) 68331 no TPU (cheaper than a 332 ) 68302 I forget... Anybody remember? The 340 has a "background debug mode" implemented in IPU microcode. Instruction execution is suspended when the 340 is in background mode. Using the debug serial I/O port you can change registers, examine memory, etc without needing a debug monitor. My only complaint about the 683xx series is that they are too expensive right now. Craig Keithley, Apple Computer keithley@applelink.apple.com [standard disclaimers apply!]
esmith@goofy.apple.com (Eric Smith) (11/15/90)
In article <11249@goofy.Apple.COM> keithley@applelink.apple.com (Craig Keithley) writes: > 68302 I forget... Anybody remember? The 68302 is based on a 68HC000 core instead of CPU32, and has three whiz-bang serial channels. > The 340 has a "background debug mode" implemented in IPU microcode. > Instruction execution is suspended when the 340 is in background mode. > Using the debug serial I/O port you can change registers, examine memory, > etc without needing a debug monitor. The 68331, 68332, and 68340 all have this feature. > My only complaint about the 683xx series is that they are too expensive > right now. That's for sure. They seem to be competing against 80186/8, 80960, and Z-180 type controllers, and though the 683xx parts are nice, I don't think they will get very many design wins unless they come up with more realistic pricing. -- Eric L. Smith Opinions expressed herein do not necessarily reflect those esmith@apple.com of my employer, friends, family, computer, or even me! :-)
wayned@wddami.spoami.com (Wayne Diener) (11/15/90)
>In article <11249@goofy.Apple.COM> keithley@applelink.apple.com (Craig Keithley) writes: >68340 doesn't have a MMU. Dual channel DMA, dual UARTS >68332 has a TPU (Timing Processing Unit - REALLY great for automotive >apps) >68331 no TPU (cheaper than a 332 ) >68302 I forget... Anybody remember? > 68302: 68000 core, one channel independent DMA, 1152 bytes on-chip dual port RAM (shared with buffers for the serial ports), three full duplex serial ports, a synchronous serial port, non-multiplexed serial interface, three timers, 6 serial DMA channels, DRAM refresh controller & some parallel I/O. -- |---------------------------------------------------------------| | // Wayne D. Diener | | // Spokane, WA | | \\ // E-mail reply to: | | \X/ To: isc-br!hawk!wddami!wayned@uunet.uu.net | |---------------------------------------------------------------|
mcmahan@netcom.UUCP (Dave Mc Mahan) (11/15/90)
In a previous article, keithley@applelink.apple.com (Craig Keithley) writes: >68340 doesn't have a MMU. Dual channel DMA, dual UARTS >68332 has a TPU (Timing Processing Unit - REALLY great for automotive >apps) It was designed due to the automotive applications market, but it is also quite useful outside that arena. I'm using it in a medical data transmission product to gather data (heart information, ECGs and such) from an implanted defibrillator made by the company I work for. The data will be sent via phone modem to the doctor for analysis. The TPU is ideal for everything from debouncing pushbuttons to generating unique pulse timing waveforms that carry PWM data to our device implanted within the body. >68331 no TPU (cheaper than a 332 ) I think this part does have a TPU, but it as only 8 seperate channels instead of the 16 channels provided by the 332. It seems that Motorola is pushing this part more heavily in the advertising (as compared to the 332) but that could be just because it is newer and needs more exposure. >68302 I forget... Anybody remember? This chip has various hardware that is useful for decoding digital telecom data lines such as ISDN, CEPT, and LAN standards. I believe it is mainly intended for connection to a high speed digital serial data line (using one of the standards) and decoding and encoding desired channels of data. >The 340 has a "background debug mode" implemented in IPU microcode. >Instruction execution is suspended when the 340 is in background mode. >Using the debug serial I/O port you can change registers, examine memory, >etc without needing a debug monitor. The 332 also has this feature. I would bet that the 331 also has it, but I don't know for sure. The documentation from Motorola "strongly recommends" that any design for the 332 include the ability to access this feature. All you really need to do is put an 8 position header on your product that makes certain pins of the CPU available in a convienient fashion. >My only complaint about the 683xx series is that they are too expensive >right now. This is quite true. I think the 68332 is down to around $180 or so now. About 8 weeks ago, it was priced at about $260 from a distributor we deal with. Of course, this $260 amount was for sample quantities of 2 so may be slightly inflated. The other complaint I have is that you have to buy a minimum of one tray. A tray is 36 parts. Quite a large cash outlay for prototype work. If you want to do volume surface mount production, a tray of parts is no big deal. For designing the product, though, it's a bit steep. I was told back in late August or early September that the 332 should cost about $30 by the end of 1Q91. We will see if they hit that target. >Craig Keithley, Apple Computer >keithley@applelink.apple.com -dave -- Dave McMahan mcmahan@netcom.uucp {apple,amdahl,claris}!netcom!mcmahan
esmith@goofy.apple.com (Eric Smith) (11/16/90)
In article <16849@netcom.UUCP> mcmahan@netcom.UUCP (Dave Mc Mahan) writes: > This is quite true. I think the 68332 is down to around $180 or so now. > About 8 weeks ago, it was priced at about $260 from a distributor we deal > with. Of course, this $260 amount was for sample quantities of 2 so may be > slightly inflated. The other complaint I have is that you have to buy a > minimum of one tray. A tray is 36 parts. Quite a large cash outlay for > prototype work. Motorola is advertising a price of $49 for the 68332, and $36 for the 68331, in quantity 1000. The single quantity price is usually not more than twice the q. 1000 price. Are you sure the $180 or $260 price is recent? -- Eric L. Smith Opinions expressed herein do not necessarily reflect those esmith@apple.com of my employer, friends, family, computer, or even me! :-)
mcmahan@netcom.UUCP (Dave Mc Mahan) (11/16/90)
In a previous article, esmith@goofy.apple.com (Eric Smith) writes: >In article <16849@netcom.UUCP> mcmahan@netcom.UUCP (Dave Mc Mahan) writes: > >> This is quite true. I think the 68332 is down to around $180 or so now. >> About 8 weeks ago, it was priced at about $260 from a distributor we deal >> with. Of course, this $260 amount was for sample quantities of 2 so may be >> slightly inflated. > >Motorola is advertising a price of $49 for the 68332, and $36 for the >68331, in quantity 1000. The single quantity price is usually not more >than twice the q. 1000 price. Are you sure the $180 or $260 price is >recent? No, I'm not sure of the numbers. I'm calling Motorola even as I type this to find out a good value. Please hold....... I just talked with the product engineer for Motorola on the phone. He was reluctant to quote a specific price, but he said he thought they were about $50 in 1000 quantity, so it sounds like your number is lots better than my original one. I need to call the local field office to get a 'real' quote. I'll let you know what I find out. >Eric L. Smith Opinions expressed herein do not necessarily reflect those >esmith@apple.com of my employer, friends, family, computer, or even me! :-) -dave -- Dave McMahan mcmahan@netcom.uucp {apple,amdahl,claris}!netcom!mcmahan
ron@motmpl.UUCP (Ron Widell) (11/17/90)
Dave Mc Mahan (mcmahan@netcom.UUCP) writes: > > It was designed due to the automotive applications market, but it is also > [...] The advantage here to the rest of the world is that automotive applications require huge volumes and low prices. So when the production volumes start, we'd better be able to build these things profitably. > > I think this part does have a TPU, but it as only 8 seperate channels instead > of the 16 channels provided by the 332. It seems that Motorola is pushing > [...] Nope, it's more like the timer in the HC11 MCU's; a couple of 16-bit free- running counters with 1 9-bit prescaler, 3 input catures, 4 output compares, 1 input cature/output compare (SW selectable), 1 pulse accumulator, 2 PWM outputs, and an optional external clock input. Much simpler than the TPU and requires < 25% (I don't remember the actual number, probably ~10%) of the die area. For those who don't need everything the TPU provides, the 68331 may be a viable, cost-effective alternative. > [...] You folks know as much as (probably more than) I do in re. current prices. I'm not authorized to talk pricing, only functionality (although, when pushed I can come up with some 'order-of-magnitude' guesses). > inflated. The other complaint I have is that you have to buy a minimum of > one tray. A tray is 36 parts. Quite a large cash outlay for prototype work. > If you want to do volume surface mount production, a tray of parts is no > big deal. For designing the product, though, it's a bit steep. I was told > [...] Our AOQ (average outgoing quality) is high enough that a significant ( > 10%) proportion of quality/reliability problems are directly attributable to the handling of devices (test, mark, package, ship). The MPQ (minimum packaging quantity) of 36 units (one tray) is an effort to minimixe the handling, thus reducing the failure rate. This, in turn, reduces our manufacturing cost and our pricing for production volumes. There is special packaging (and a special part number) which allows you to get these parts modulo 2, rather than modulo 36. You do, however pay at the single-piece pricing rate. I apologize if this just sounds like a sales pitch, there are times when knowing the reasons for a policy make it easier to understand, accept, or tolerate. Regards, -- Ron Widell, Field Applications Eng. |UUCP: {...}mcdchg!motmpl!ron Motorola Semiconductor Products, Inc., |Voice:(612)941-6800 9600 W. 76th St., Suite G | I'm from Silicon Tundra, Eden Prairie, Mn. 55344 -3718 | what could I know?
mcmahan@netcom.UUCP (Dave Mc Mahan) (11/17/90)
In a previous article, ron@motmpl.UUCP (Ron Widell) writes: >Dave Mc Mahan (mcmahan@netcom.UUCP) writes: >> >> It was designed due to the automotive applications market, but it is also >> [...] >The advantage here to the rest of the world is that automotive applications >require huge volumes and low prices. So when the production volumes start, >we'd better be able to build these things profitably. I guess the big question is: when are production volumes going to start? I agree that standardizing in the auto market will insure large enough sales to support the making of the part, but how long will it be before the chip is stable and I can get manuals that aren't marked 'preliminary' and have lots of typos and errors? This is a great part, but I'd like to know how soon it will be stable. Also, the SIM manual states that the part will be produced in a plastic package, but all I've seen so far is ceramic. When is the package going to change over to the plastic version? >> The other complaint I have is that you have to buy a minimum of >> one tray. A tray is 36 parts. Quite a large cash outlay for prototype work. >> If you want to do volume surface mount production, a tray of parts is no >> big deal. For designing the product, though, it's a bit steep. >Our AOQ (average outgoing quality) is high enough that a significant ( > 10%) >proportion of quality/reliability problems are directly attributable to the >handling of devices (test, mark, package, ship). The MPQ (minimum packaging >quantity) of 36 units (one tray) is an effort to minimixe the handling, thus >reducing the failure rate. This reduces our manufacturing cost and our >pricing for production volumes. There is special packaging (and a special part >number) which allows you to get these parts modulo 2, rather than modulo 36. >You do, however pay at the single-piece pricing rate. I have purchased about 10 of the parts in the little two unit plastic boxes for doing our proto run. I like the plastic package concept you came up with, it looks like a very functional way protect all those sensitive little leads. Will the part ever be made in a PGA form? This would make proto development mucho easier to do. Changing to a new die rev would be much simpler than cutting off the current chip, desoldering 132 little dangling legs by hand, and then soldering on the new improved chip. Another question I have that I can't seem to get answered by the product engineer in Austin is: I'm currently using the BCC (business card computer) that comes with a 332, some RAM, some EPROM, and a crystal on it. (nice little scheme, by the way!) The documentation states that I can re-program the EPROM on board by erasing the old one and then using the BCCDI (business card computer development interface) to send over the new code. Ok, I like the idea and the way it is done, but how do I get the combination to generate the proper strobe signals for writing to the EPROM? From the book I have on EPROMs, you have to hold the chip-select active for something like 1 millisecond or so while you assert the data lines and the write signals that cause the EPROM to program the desired data. Since the chip-select in question is tied directly to the 332 CPU chip, how can I hold it active long enough without blowing up the 332? Nobody I can find in Austin seems to know. I'm looking for nitty-gritty, technical, detailed steps for a way to accomplish this task. >I apologize if this just sounds like a sales pitch, there are times when >knowing the reasons for a policy make it easier to understand, accept, or >tolerate. Don't worry about the pitch. I understand why the parts come that way, I'm just wishing for something that suites my needs (those of the little guy in a niche market) better. >Ron Widell, Field Applications Eng. |UUCP: {...}mcdchg!motmpl!ron >Motorola Semiconductor Products, Inc., |Voice:(612)941-6800 -dave -- Dave McMahan mcmahan@netcom.uucp {apple,amdahl,claris}!netcom!mcmahan