jdg@elmgate.UUCP (Jeff Gortatowsky) (06/05/86)
Much to my suprise I recieved 1.2 beta2 yesterday! Guess I *WAS* one of the
1000! Anyways I quickly got some blanks and started making a working set
of disks. First problem:
The tools disk was badly damaged. Diskcopy said NO_WAY_BUDDY!!!
Thanks to a program called disksalvage (not diskdoctor) I was able to
recover most of the executables. However most of the sources directory
was cut. The first two tracks of the disk were completly gone. Odviously
diskdoctor expects a formatted disk and then just fixes up the
filesystem. It won't play if the media is munged up. Disksalvage on the
other hand copies every readable block to a new formatted disk then
fixes the filesystem.
Question to CBM-Amiga:
What number is the right number to call to get a replacement (free or
otherwise)? The bug report sheet has an address but no phone number.
I won't send the disk back. The magnetic media is physically damaged.
Tho I'll send a fresh one along if needed. The number please.....
Didn't get a chance to read through (or even print) the docs on disk yet. A
quick browse revealed wierd spacing (meaning none in some cases) between
words. I'll have to see if they are to printed by some special program.
The emacs was fine but (sniff) the docs were on one of the two bad
cylinders.
I'd say directories are twice as fast. No more no less. At least with
default settings. Moving icons around is much nicer.
As reported earlier Manx exec's fail (double sniff). Could this in part be
due to the new path support? I saw the exact same message from make under
1.1 when 'make' could not find a command I had renamed (DELETE to rm).
I guess I'd better read the new docs first before jumping to conclusions.
I did find a little 'glitch' in the systems error handling. My workbench
(1.2) copy had a 'marginal' block on it. From the CLI I typed 'dir opt a',
when the system got to the questionable block in retried twice and put up a
system requester telling me about the read error. When I clicked (or Amiga
keyed) cancel the requester went away and there I sat. The dir command
did not abort nor did it continue. It just sat there until I keyed CONTROL
'c'. Then I got back my CLI prompt. It was repeatable. Bug report is on
the way as soon as I make acouple of copies of the blank provided.
Metascope seemed to work fine.
Wack is as undocumented as ever. The font editor is superb. Someone should
be patting him or herself on the back. Toggle switches, I love it. Freemap
died. Just puts up a screen of garbage. Fortunatly the quit menu was still
working. Don't know if this is a bug or just an 'unsupported' program.
Can't look at the source. If it was there, it was on one of the dead
cylinders.
With Manx only kinda working I'll have to dust off the Lattice disk to try
some of the new tricks (ick). Now let me see.. which disk is that PD make
on? Oh well, guess I can always hack on the ST while the linker is running,
aye? 8-)
The PD shell version 2 also seemed to work fine.
More as I find it......
Jeff
PS> RJ Mical! You back for good? Or just visiting/consulting? Either way,
good to see your comments again!
--
Jeff Gortatowsky {allegra,seismo}!rochester!kodak!elmgate!jdg
Eastman Kodak Company
<Kodak won't be responsible for the above comments, only those below>daveh@cbmvax.cbm.UUCP (Dave Haynie) (06/06/86)
> The tools disk was badly damaged. Diskcopy said NO_WAY_BUDDY!!! > Thanks to a program called disksalvage (not diskdoctor) I was able to > recover most of the executables. However most of the sources directory > was cut. The first two tracks of the disk were completly gone. Odviously > diskdoctor expects a formatted disk and then just fixes up the > filesystem. It won't play if the media is munged up. Disksalvage on the > other hand copies every readable block to a new formatted disk then > fixes the filesystem. > -- > Jeff Gortatowsky {allegra,seismo}!rochester!kodak!elmgate!jdg > Eastman Kodak Company > <Kodak won't be responsible for the above comments, only those below> Is that DiskSalv? I wrote DiskSalv, and put it up on the net awhile ago. I'd always try recovering a disk using DiskSalv first. It does nothing at all to your original disk, so you can always try something else later. The DiskDoctor program is supposed to try to rebuild the bad disk itself, so it has the potential at least of damaging a badly trashed disk. The last version of DiskDoctor I saw would sometimes make a dangerous recovery; i.e., the files were recovered but anything new put on the disk could cause it to get clobbered again. DiskSalv rebuilds the new disk at the DOS level, it doesn't try to play around with rebuilding a disk's structure, so with it you'll always have a safe disk as a result, though there certainly are things DiskSalv can't restore (basically any disk block that's pronounced unreadable by the trackdisk.device, though a directory entry on one of these blocks has no effect on the ability to recover an otherwise OK file member of that directory). -- /\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\ Dave Haynie {caip,inhp4,allegra,seismo}!cbmvax!daveh "I read dozens of books, 'bout heros and crooks, and I learned much, of both of their style.." -Jimmy Buffet These opinions are my own, though for a small fee they be yours too. \/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/
rj@amiga.UUCP (Robert J. Mical) (06/07/86)
Keywords:
Thanks! I'm back for good, if I can help it! RJ >:-{)*jdg@elmgate.UUCP (Jeff Gortatowsky) (06/09/86)
In article <357@cbmvax.cbmvax.cbm.UUCP>, daveh@cbmvax.cbm.UUCP (Dave Haynie) writes: > > Is that DiskSalv? I wrote DiskSalv, and put it up on the net awhile ago. > I'd always try recovering a disk using DiskSalv first. > -- > /\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\ > Dave Haynie {caip,inhp4,allegra,seismo}!cbmvax!daveh > \/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/ Yes. It was disksalv. I got it off of CIS. I musta missed the posting here on the NET. As soon as I realized Disked wasn't gonna hack it, I logged on to CIS and dl'd it. I didn't even pay attention to who the author was. Thanks Dave. Good tool. Invaluable. Credit where credit is due, nice job Dave. From an unusable disk to a usable in nothing flat. Still all the source is gone. Nothing that any program could have done about that. It just wasn't there! My (and everyone elses for that matter) 1.2 beta2 has got alot more serious problems with it than bad distribution disks however. It was not ready for beta testing. No way, no how. Sure it's nice to use your developers as QA people and to receive feedback. But my experiance with 1.2 has been negative enough that I'm considering going back to 1.1 to continue development until I get a version of 1.2xx that at least works as well as 1.1. Or at least doesn't cause new bugs in things that did work before. Since another person thinks I'll save someone else the trouble of finding bugs I've already found I'll expand alittle. I've only used 1.2 for a weekend. One day really. I stress that all I did was play with C-A's stuff. I didn't get around to programming anything yet. Disk errors. As I stated in a previous posting, my tools disk was bad screwed up. Two unreadable tracks. In the CLI I recovered what I could and reformatted after going back to the Workbench (WB). Diskcopy then reported error as I tried to use that disk (format said it was fine). I clicked <CANCEL> in the system requester that appeared after diskcopy aborted. It retried! After 20 times I gave up and Ctl-Amiga-amiga'd the thing. Using disked I munged up a disk on purpose. Same thing. Had to reboot. In others words error recovery from a disk error is reboot. Try this. Make a copy of the original 1.2beta workbench (need this so when you open the disk icon the window will be in the 'default' screen position). Now write protect this copy. Place it in df0: and reboot. Open the disk icon (named copy 2 of WB xxxxx). Now select the disk icon again and try renaming it. When the requester comes up saying the disk is write protected press cancel. Poof! Where's the open window's close gadget and upper scroll gadget go? Retry the rename operation. When the string gadget comes up click outside of the string box. Multi-tasking? What multitasking? While your there staring at the sleepy cloud go ahead and move the open window. What if I needed to see something (anything) underneath that window before I decided what I want to rename this disk to? Try Open or Save As.. from the notepad. Permanent filename requester. Time for Ctrl-Amiga-Amiga. Try word wrap. Yup it wraps. But leaves the image of the word on the previous line. (BTW all this was done with the notepad window openned up full screen). How come the scroll arrows redraw the screen (no they didn't scroll). Printing was fun. But before I say it's buggy I better hack up a driver for my Star SG15 that I know works. Right now I use the Star in the IBM mode and the MPS-1000 driver as per CBM's recommendation. But I don't KNOW that's what it should be. Printing put some line breaks in where there were none, didn't put breaks (ie. LOOONG lines) where there were short lines etc... I don't know,... it *USE* to work under 1.1. Forget about all the neat stuff for programmers for a moment. This is stuff the USER (ie. guy who buys the programs we write) is going to do. This is the stuff that MUST BE ROCK SOLID!!!!!! THIS IS THE PART OF THE AMIGA THAT MUST BE INTUITIVE!! Remember? Intuition. It must work regularly and predictabily EVERYTIME. No matter what the user does. I know, I know, alot easier said than done. So the solution. Forget about my and other programmers wish list for the moment. Concentrate ALL efforts on bomb proofing the USER's view of AmigaDOS/Intuition as supplied. We programmers will live without for a while. What WE need to sell our wares is USER's. Preferably a user who is not frustrated everytime he sits down at the machine. Before you go flame me to a crisp. I DO REALIZE this is beta stuff. I just think it should be ALPHA. Further it looks like they paid to much attention to what we, the programmers wanted and not enough to honing and perfecting what they already had. BTW. What this programmer really needs is a stable environment. So that I know at least 95% of whatever goes BOOM in my code is MY code. Sorry once again for the size of this posting. Flame away..... -- Jeff Gortatowsky {allegra,seismo}!rochester!kodak!elmgate!jdg Eastman Kodak Company <Kodak won't be responsible for the above comments, only those below>
jimm@amiga.UUCP (Jim Mackraz) (06/10/86)
In article <469@elmgate.UUCP> jdg@elmgate.UUCP (Jeff Gortatowsky) writes: > >My (and everyone elses for that matter) 1.2 beta2 has got alot more >serious problems with it than bad distribution disks however. It was not >ready for beta testing. No way, no how. Sure it's nice to use your >developers as QA people and to receive feedback. But my experiance with 1.2 >has been negative enough that I'm considering going back to 1.1 to continue >development until I get a version of 1.2xx that at least works as well as >1.1. Or at least doesn't cause new bugs in things that did work before. > Let me make it very clear for you. You ARE in fact the rigorous testers. The so-called alpha phase consisted of a small release to developers, the so-called beta phase consists of a wider distribution. We are depending on developers to be exercising these releases. If you are interested in a solid 1.2 system release, you will continue to use and exercise the beta releases. Many people are doing so, reporting the bugs they find, and claim to be generally happy. You were sent information on how to report bugs. We try to make sense out of problems posted to usenet, but you really should send them directly to us. To report bugs: Nancy Rains Commodore-Amiga, Inc. 983 University Ave. Bldg D. Los Gatos, CA. 95030 or ...pyramid!amiga!nancy > >Before you go flame me to a crisp. I DO REALIZE this is beta stuff. I just >think it should be ALPHA. Further it looks like they paid to much >attention to what we, the programmers wanted and not enough to honing and >perfecting what they already had. > I don't think this newsgroup deserves a long discussion on the naming of different test phases. There hasn't been a lot of time to worry about the public relations aspects of naming this release. In my experience, alpha and beta had more traditional meanings, to be sure. >Flame away..... Cool your jets. jimm
higgin@cbmvax.UUCP (06/12/86)
In article <469@elmgate.UUCP> jdg@elmgate.UUCP (Jeff Gortatowsky) writes: >... >My 1.2 beta2 has got alot more >serious problems with it than bad distribution disks however. It was not >ready for beta testing. ... But my experiance with 1.2 >has been negative enough that I'm considering going back to 1.1 to continue >development until I get a version of 1.2xx that at least works as well as >1.1. Or at least doesn't cause new bugs in things that did work before. > My suggestion: develop under 1.1 and test under 1.2 if you REALLY need the new functionality in the 1.2 OS. As a Manx C user, I HAVE to continue developing under 1.1 because it won't run on 1.2 (Manx's fault, not C-A's). More seriously though, you should ask yourself whether 1.2 will be in USERS's hands (i.e., officially released and available in stores) when your commerciual product will be available. It is my guess that 1.2 won't be available in STORES until about November. This is purely my opinion - I might be way off base. > <...numerous valid gripes on using 1.2 from a user's standpoint...> > >Forget about all the neat stuff for programmers for a moment. This is stuff >the USER (ie. guy who buys the programs we write) is going to do. This is >the stuff that MUST BE ROCK SOLID!!!!!! THIS IS THE PART OF THE AMIGA THAT >MUST BE INTUITIVE!! Remember? Intuition. It must work regularly and >predictabily EVERYTIME. No matter what the user does. This I don't necessarily agree with. You're correct in saying the user must never see an OS's bugs, but C-A is trying to do its best to see that they respond to developer's needs, wishes, and whims, which is refreshing. The 'user' problems are a given - I'm sure C-A knows what they have to fix up, and that's why beta1.2 goes to developers and not users - to those who NEED the enhancements which are in response to their needs, wishes and whims. >So the solution. Forget about my and other >programmers wish list for the moment. Concentrate ALL efforts on bomb >proofing the USER's view of AmigaDOS/Intuition as supplied. We programmers >will live without for a while. What WE need to sell our wares is USER's. >Preferably a user who is not frustrated everytime he sits down at the >machine. > Ah, but what Commodore needs is good software, so it must take care of the developers first! You can't have users without good software, and as you said, 1.1 is pretty solid, and so user's won't be complaining right now about the user-interface - they will be SCREAMING for Lotus-123, Excel, Word, Symphony, dBase, etc!!!! >BTW. What this programmer really needs is a stable environment. So that I >know at least 95% of whatever goes BOOM in my code is MY code. >Sorry once again for the size of this posting. > I'm sure using developing with CLI and DOS like usual is still pretty ok under 1.2. There are people at CBM doing it, I know. Maybe it's just workbench that has problems. Regards, Paul Higginbottom Disclaimer: I do not work for Commodore and my opinions are at least my own.