This message contains an entire topic ported from a separate forum. The original message and all replies are in this single message. We have seeded this new forum with selected information that we expect will be of value to you as you search for answers to your questions.
Posted: Sun Feb 13, 2005 10:07 pm
Nothing is perfect, is it. so what would you change about the HC08 or S08, features, specs, anything etc.
Posted: Mon Feb 14, 2005 9:19 am
Nope, nothing is perfect, you are right.
As a heavy user of HC08, I would like spec to be more similar. I mean some info were removed from new spec and you need to go to old ones to know (hysteresis value on I/O are never specified, but the the AZ spec gives an idea).
I would also like to have some kind of module version in the family. Many modules are changed from one MCU to another. I understand that it is for the best, but it's time consuming to check if a 'coma' has changed between two CGM modules. If I had CGM v2.0 and CGM v2.3, I would know I can't necessarily apply what I knew...
Posted: Tue Feb 15, 2005 8:40 pm
As a heavy user of HC08, I would like spec to be more similar. I mean some info were removed from new spec and you need to go to old ones to know (hysteresis value on I/O are never specified, but the the AZ spec gives an idea).
I would also like to have some kind of module version in the family. Many modules are changed from one MCU to another. I understand that it is for the best, but it's time consuming to check if a 'coma' has changed between two CGM modules. If I had CGM v2.0 and CGM v2.3, I would know I can't necessarily apply what I knew...
Posted: Tue Feb 15, 2005 8:40 pm
Hmmm.... I have been thinking of that.... Would improve TIM and add another free running counter to have independent timing ready from the PWM feature... I would add some HW config support to handle outputs on input change with no direct SW intervention. Just configuration of the action and input trigger condition in advance and go doing sth else. Current solution is quite cumbersome, interrupt handling, set action, return handling... Toooo long... Ufff! I would add 32kHz CMG PLL module to each device and increase clock at least 4x and finaly improve the "oldie" CPU core as Toshiba did with 6803... Much less clocks per instruction... Hmmm... Sometimes I think, Freescale takes after the company I work for. 
Too big, too slow, too cumbersome... Hmmm, markets are happy, nobody needs more, why do we need change... Anyway, I like good old Motorola/Freescale...
Too big, too slow, too cumbersome... Hmmm, markets are happy, nobody needs more, why do we need change... Anyway, I like good old Motorola/Freescale...Posted: Wed Feb 16, 2005 8:15 pm
what do you think about the S08 vs the HC08, do you want even more performance?
Posted: Thu Feb 17, 2005 9:48 am
Well, many beautiful features like BDM, EMC slew rate control, KBI with edge polarity control and so on... Many features the competitors like ATMEL, MICROCHIP and others could only dream of. But, the CPU itself is another story. If we say, that an averagy cycle count per instruction is approx. 4, then we can get 5 MIPS. Not very stunnig result for 20MHz clock. The best competition usually offers sth around 15-20 MIPS. A good example could be Rjindael AES implementation on 8051/AVR and HC08 platform. Fortunately, majority of applications are happy with this and take advantage of another excellent features. I came up with this story: DC current motor regulation would require 100mV/7-9bits. It was impossible to do it when the VREF pins have to be tided up to VADD-VDD. Additional analog network have to added... I would say, that HC08, HC11, HC12 stuffs are very appropriate for applications where any kind of math computation is needed, however, they are mostly excluded from fast response logic automats kind of application especially due to long interrupt latency, long timing BSET/BCLR, missing skip instruction. As I have said before, the TPM and TIM modules are not designed perfectly. It is quite tricky setting let say PWM for 40kHz and HC11 RTI analogy for 10ms regular interrupt...
We can discuss another day and night about bla-bla..... Anyway, HCS08 is a great product for this class, I think from HW and features point of view it is the best. But the CPU performance falls behind of expectations, at least mine.
We can discuss another day and night about bla-bla..... Anyway, HCS08 is a great product for this class, I think from HW and features point of view it is the best. But the CPU performance falls behind of expectations, at least mine.
