|
|
View previous topic :: View next topic |
Author |
Message |
Jeff King
Joined: 20 Oct 2003 Posts: 43 Location: Hillsdale, Michigan USA
|
ICDU-40 slower then ICD2? |
Posted: Wed Jan 07, 2004 2:51 pm |
|
|
Hi:
I got an ICDU-40 because of the claims of faster speed then the ICD2. However, I am not observing this. These are the results I obtained, uploading a program to a 18F8621, about 50% full:
ICDU-40: ......................................... 55 seconds
ICD2 "Let ICD2 chose settings": .... 18 seconds
ICD2 "Program everything": ........... 37 seconds
PCWH was 3.182 and MPLAB was 6.40
Now, carefully reading the wording at:
http://www.ccsinfo.com/faq/?46
we see they claim a speedup in single stepping, comparing a ICD2 to a ICDU-40, of a greater then 5 speedup. Suffice to say, I assumed it would be at least as fast as the ICD2 in program upload, or even faster, but I am seeing the upload speed REDUCED by a factor of 3.
I'm hoping I'm doing something wrong, otherwise this is unacceptable as the few seconds I save in single stepping is more then offset by the 3X increase in upload times.
Anyone know if this is a known problem or am I missing something?
Thanks
-Jeff
PS
My ICDU-40 also had the RJ-45 problem Todd reported some time back, bending the pins fixed it but it sure is scary.... |
|
|
Neutone
Joined: 08 Sep 2003 Posts: 839 Location: Houston
|
|
Posted: Wed Jan 07, 2004 4:17 pm |
|
|
Is your ICD-U got the latest firmware? It is a seperate download. That page full of speeds is a bit old. Both are faster now. |
|
|
Jeff King
Joined: 20 Oct 2003 Posts: 43 Location: Hillsdale, Michigan USA
|
|
Posted: Wed Jan 07, 2004 6:01 pm |
|
|
Neutone wrote: | Is your ICD-U got the latest firmware? It is a separate download. That page full of speeds is a bit old. Both are faster now. |
ICDU firmware is from December 22nd, 2003 , version 1.20 #97. It is the most current that is on the CCS site.
Don't believe me, run your own test if you have a ICD2. There is a dramatic difference in the upload times.
Here is what I got from CCS when I reported this problem:
--- Original Message ---
From: "CCS Support" <[email protected]>
To: ?<[email protected]>
Cc:
Sent: Wed, 7 Jan 2004 16:43:25 -0600
Subject: Re: ICDu-40 slower then ICD2?
?
>This message is a reply to CCS e-mail id: 4A2812
>
>We have not really worked much on the download time as our initial
>concern was debug time. ?We are now working on a major revision to
>the ICD firmware that will speed up the download time. ?This is
>expected out in about a month.
--------------------------------
The (apparent) implication this device is faster then the ICD2 on the web page(s) really has me disappointed and wasted alot of my time. I hope they fix it but in the future, full disclosure would really be appreciated in my future dealings with them. I have egg on my face now.
Last edited by Jeff King on Wed Jan 07, 2004 9:13 pm; edited 1 time in total |
|
|
Neutone
Joined: 08 Sep 2003 Posts: 839 Location: Houston
|
|
Posted: Wed Jan 07, 2004 6:56 pm |
|
|
Moe: Hey Homer check out my new deep fat fryer. I can do an entire cow in 15 seconds.
Homer: But I'm hungry now. |
|
|
Mark
Joined: 07 Sep 2003 Posts: 2838 Location: Atlanta, GA
|
|
Posted: Wed Jan 07, 2004 7:44 pm |
|
|
You shouldn't need to "download" (program) the chip that often. If you are programming that often, then.... well.... I guess your programming skills are lacking. |
|
|
Guest
|
|
Posted: Wed Jan 07, 2004 7:47 pm |
|
|
Neutone wrote: | Moe: Hey Homer check out my new deep fat fryer. I can do an entire cow in 15 seconds.
Homer: But I'm hungry now. |
Sorry, with the mad cow scare I'm confused about your bovine analogies...
Maybe state it another way... |
|
|
Jeff King
Joined: 20 Oct 2003 Posts: 43 Location: Hillsdale, Michigan USA
|
|
Posted: Wed Jan 07, 2004 7:55 pm |
|
|
Whatever.
Last edited by Jeff King on Wed Jan 07, 2004 9:18 pm; edited 2 times in total |
|
|
Mark
Joined: 07 Sep 2003 Posts: 2838 Location: Atlanta, GA
|
|
Posted: Wed Jan 07, 2004 7:58 pm |
|
|
Nope, you are just whinning. CCS already sent you an email stating they were working on the download speed! |
|
|
Jeff King
Joined: 20 Oct 2003 Posts: 43 Location: Hillsdale, Michigan USA
|
|
Posted: Thu Jan 08, 2004 1:32 am |
|
|
Mark wrote: | Nope, you are just whinning. CCS already sent you an email stating they were working on the download speed! |
Did that buy me back the 3-4 hours I wasted on it due to their lack of disclosure in my pre-sales inquires? Did they change their web page to indicate the product is currently slower then the ICD2 on program uploads? You see, I do this for a living, and wasted time to me is wasted money. Even with the slower speed, the product is a good deal, but those going from the ICD2 to the current rev of the product will be disappointed if their code sizes are large.
More to the point, this board is about sharing information, and that is exactly what I did. Another CCS user shared the information about the bad RJ-45's on the ICDU's, and that saved me alot of time, when I had the exact same problem yesterday. I was just trying to return the favor, and save someone elses wasted time.
So, the fact you call this "whining" and hurl insults means little to me. It is useful factual information I posted. And rest assured, when CCS fixes this (I'm sure they will), I'll run the test again and post the results.
Key words: ICDU-40, slow, upload, Version 1.20, ICD2 |
|
|
Guest
|
|
Posted: Thu Jan 08, 2004 2:43 am |
|
|
Mark I'm surprised to see you posting this way. I thought some time ago that you were trying to be the next "PCM Programmer" but you fail to see that he is strictly professional and never insulting.
The bottom line... the ICD-U40 is apparently slower than implied. |
|
|
Mark
Joined: 07 Sep 2003 Posts: 2838 Location: Atlanta, GA
|
|
Posted: Thu Jan 08, 2004 7:43 am |
|
|
I would rather have faster single stepping than faster programming for debugging! If you are wasting so much time because you change a line of code, compile, download, and then run, then I don't think the problem is the download speed. Yeah it might be a pain because you do it so much but the fact is, the person's programming skills are lacking. If such a person didn't have to compile so much, then their time spent would be greatly reduced. I wasn't throwing insults, I was merely stating facts in the nicest way that I could come up with at the time. The link that Jeff gave clearing states the debugging time (single stepping) and says nothing on programming speed. Now about the whinning statement, I don't feel that I am insulting my 4 year old when I tell her to quit whinning nor the 37 year old that sits across from me at work CCS already told you that they were working on that. But that is not good enough for you. Did you ever ask what the programming speed was? If that was of great concern then you should have asked for that specification. I don't see how they cost you the 3-4 hours of wasted time. If you spent 3-4 hours programming a chip for debugging code, then there is a problem but I am not going to say what it is.
I apologize for making you feel insulted if your only intentions are posting the facts. That is a good thing. I guess I would have worded it a little differently.
PS: If you are that concerned with the programming time and it does waste a lot of your valuable time, I would look into getting an emulator. They are much faster especially for the compile/run/test compile/run/test type of programming.
PSS: I actually do this sometimes when making lots of changes to existing code to make sure that if something "breaks" it is easy to find! |
|
|
Guest
|
|
Posted: Thu Jan 08, 2004 1:02 pm |
|
|
I don't want to make whatever gripe you have with Jeff worse by letting you think that I am him. I'm not.
My apologies Jeff.
PS The word I used was "implied" and to advertise faster single-stepping would imply faster..... everything. CCS does address issues and I'm sure they'll work it out. |
|
|
Jeff King
Joined: 20 Oct 2003 Posts: 43 Location: Hillsdale, Michigan USA
|
|
Posted: Thu Jan 08, 2004 3:01 pm |
|
|
Anonymous wrote: | I don't want to make whatever gripe you have with Jeff worse by letting you think that I am him. I'm not.
My apologies Jeff. |
No need for it, understand the mixup with 'guest' as I did it once when I forgot to log in (the homer and the cow clarification request was mine).
But I am suprised about this rabid reaction..... it is like they are treating me like Trampas or some other anti-CCS shill. I found a problem, posted a query about the problem, confirmed the problem existed, and posted the facts of the problem to the group to share my experince.
As far as Mark's critism of my programming, which he has never seen, we each code our particular ways for the project at hand, and in my case, working with the IOSOFT 802.11b package, which compiled is 32kb and all the settings hardcoded, faster uploads are a big deal, or in this case, not being 3 times slower then the ICD2. Working with a IRQ driven program this size is far far different then a 500 byte program. Most of the time, if I have a problem, I set a breakpoint and look at the results in the watch window to see if they make sense. Reason I don't single step that much is I know what my code is supposed to do, and don't need to run along side it to make sure it does what it is supposed to do. It is quicker for me to break at a known point, and exam all my registers to make sure they are what I think they should be. So, to me, single step time is not that big of a deal as I stopped using it to a large degree years ago. But that may have to do with the fact I mostly do real time, IRQ driven stuff, like radio interfaces and serial interfaces. Single stepping just can't work here.
I am a long time CCS customer and am sticking with the compiler product. But just because of this, I don't think they walk on water and do make mistakes. They made a mistake in their advertising by implying the ICDU-40 was faster then the ICD2, and I'm sharing that information here so others don't make the same assumption I did. Until they get the product on parity with the ICD2, at a minimum, they should remove all the comparisons with the ICD2. The product is cheap enough that they still will sell a ton of them without needing to market it in this manner.
And yes, I think they will address this, and I'll post the results when they fix it.
Thanks (whoever you are)
-Jeff |
|
|
Neutone
Joined: 08 Sep 2003 Posts: 839 Location: Houston
|
|
|
Darren Rook
Joined: 06 Sep 2003 Posts: 287 Location: Milwaukee, WI
|
|
Posted: Thu Jan 08, 2004 4:42 pm |
|
|
Jeff King wrote: |
it is like they are treating me like Trampas.
|
Ahhh... This brings back fond memories..
Honestly, I didn't know about the slow programming time either. I can't wait for the faster firmware too - I do a lot of big programs in the PIC18. |
|
|
|
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
|
Powered by phpBB © 2001, 2005 phpBB Group
|