CompuServe Thread

#2.0 ROM Release Info

22 messages in this thread
#13882From: Larry FeldmanAug 28, 1991 8:57 AM
The following is a statement from Bill Sydnes, Commodore's VP of Engineering, regarding the status of the 2.0 release of the Amiga operating system: As part of Commodore's ongoing commitment to our customers, I am pleased to announce completion of Release 2 of the Amiga Operating System. Today's Amiga Release 2 is very different from the Release 2 first shipped with the Amiga 3000. Many additional man-years of effort have been invested to create a polished, stable, compatible, tested release that Commodore will build on for the future. Several hundred official beta testers provided feedback, resulting in the best tested, most trouble-free and reliable operating system for the Amiga ever. For existing Amiga 3000 owners, Commodore plans to release a five disk update set. For A500/A2000 owners, Commodore will make available a Release 2 Enhancer Kit, with new ROM, diskettes and full documentation. Installation will be through an Authorized Commodore Service Center. Release 2 Enhancer Kits will be available beginning in October. Pricing will be announced shortly. Following shipment of Release 2 Enhancer Kits, Commodore will gradually phase Release 2 into new unit production. Release 2 is a major step forward in software technology, and is part of Commodore's commitment to keeping the Amiga competitive, and looking to the future. William Sydnes Vice President – Engineering, Commodore International Limited
#13910From: Jim Shaffer Jr.Aug 29, 1991 12:02 AM
Naturally I have some comments on this 🙂 (1) Are they going to release ROMs for A3000 owners, or not? (Or will they be the same as the A2000 ROMs?) (Whether I actually care about this depends on how much stuff I still have to use 1.3 for. 2.04 is supposed to have a lot of work-arounds that aren't in 2.03.) (2) Wouldn't it be smarter to put the ROMs into new units FIRST?
#13917From: BILL LEACHAug 29, 1991 12:46 PM
Jim; It is likely that ROMs will be released for the A3000 but it is also likely that it will be a "service center installation" upgrade. The ROMs will undoubtedly become a "replacement part" also, so one should be able to order the part and make the replacement. The ROMs will probably be put into new units soon. I am not personally sure that there is much to be gained by putting the ROM into a 3000. After all, the cold boot time for a 3000 is already pretty quick. The money would probably be better invested in read/write memory. Developers, OTOH, will likely want the ROM but only because that is what will actually be installed in new machines (at some future data). Usual Disclaimer applies. My guesses on this stuff is no better than anyone elses and probably not as good as the opinion of some. An Amiga message processed by Whap!, bill
#15751From: Tony PitmanOct 24, 1991 3:05 PM
In the statement it is said that the A3000 owners (i.e. me for example) will receive the upgrade on diskettes. Last time I was talking to a commodore tech (2 months ago) he said that currently the A3000 does not have the 2.0 ROMS in it. If this is going to be a ROM release, am I going to get ROMS for my A3000 or just diskettes? Tony
#15760From: Don Curtis/SYSOPOct 25, 1991 1:11 AM
Tony, 2.04 (version 37.175) is currently available on disk for the A3000 and it is the same version that will be put into ROM. Don
#15762From: BILL LEACHOct 25, 1991 10:04 AM
Tony; The ROM upgrade for the A3000 has not been released. A Commodore New Release (nwrl.03.0991) states that the A3000 ROM's (with new manual) will be available in November for $45 as product AS314. -bill/ARS KB7LX
#15947From: Tony PitmanNov 1, 1991 7:03 PM
thanks for the info. Tony
#15905From: John ToebesOct 30, 1991 10:49 PM
Actually, it is to your advantage to continue to boot the Kickstart from the hard drive. While putting the kickstart in rom will eliminate the 3 second delay in booting and does give you back the extra 1/2 meg of ram, you loose a bit of performance and more importantly the ability to boot any newer versions of 2.x that will come to pass. The Roms that are in the 3000 are actually a version of an early 2.0 alpha that has some special boot code in it. With the final release, this ability is no longer there. John A. Toebes, VIII – via Whap!
#15946From: Tony PitmanNov 1, 1991 7:03 PM
You know what would be nice? If they would put a start-up routine that copied the ROMs into RAM at boot time. This way you would get the best of both worlds. Tony.
#16006From: Jonathan HirschmanNov 3, 1991 11:58 PM
Tony– SetCPU will copy the ROMs into RAM, if * You have a MMU in your machine * You give it the correct command parameters Been doing this under 1.3 for a short while. Jonathan Jonathan – from Albany, NY, via whap!
#16021From: Tony PitmanNov 4, 1991 6:21 PM
thanks, do you know if the 2.0 ROMs are avail for the 3000 yet? tony
#16190From: Jonathan HirschmanNov 10, 1991 2:18 AM
Tony– As far as I know, nope. I've had a 3000 since the August before last, and I've yet to hear from my dealer (who has been holding a P.O. for them for some months now). Expect they'll arrive this year….:) Jonathan Jonathan – from Albany, NY, via whap!
#16632From: Tony PitmanNov 18, 1991 6:06 PM
Let me know when you hear something. Also see my note called 2.04 upgrade and tell me what you know.
#16013From: Andy FinkelNov 4, 1991 10:05 AM
You can use the CPU command to do this if you are on an MMU equipped machine. andy
#16016From: Gregory B. TibbsNov 4, 1991 3:05 PM
You can use CPU if it is a non-68040 MMU equipped machine. -Greg
#16019From: Andy FinkelNov 4, 1991 5:28 PM
You're right. And, if setcpu hasn't been updated to 040 yet, well, the source is available, and so could be modified. andy
#16040From: Gregory B. TibbsNov 5, 1991 7:46 AM
Actually, I have a working FASTROM program for the Progressive A3000 68040 accelerator . It is only marginally faster than stock with the 32 bit roms I have. Turn caches off and it is 30% faster. To anyone who has an 040: If you send me the type of machine, where your 32 bit ram is relocated, I can easily modify my MMU tables to support your board. I plan on uploading it within the week. -Greg
#16066From: Jim SakladNov 5, 1991 10:34 PM
Greg – I'm a hardware junkie, but not a hardware guru; that is, I really like new hardware and what it can do, but I do not always understand all of the ramifications. > I have a working FASTROM program for the Progressive A3000 68040 > accelerator . It is only marginally faster than stock with the 32 bit > roms I have. Turn caches off and it is 30% faster. It is possible to explain simply WHY it is faster with the caches turned off? I thought that they were there to speed up processing! . . . . . JimDoc
#16078From: Gregory B. TibbsNov 6, 1991 6:42 AM
I'm talking relative. With the caches off, the fastrom program retains much of the speed of the machine because the roms are in 32 bit ram. Thus relatively, you see 25 to 30 percent increase over the speed of the machine with the caches off.The machine overall is still slower than when the caches are enabled. (Clear as Mud, right?) When the caches are turned on, the 68040 executes much of the ROM code from the caches and thus there is little performance gained (1 – 2 %) by putting the rom code in 32 bit RAM, but it is still a gain. Giving up 512K may seem a lot for 1 – 2 percent inprovement (A2000s running from 16 bit rom will see more – I'm talking relative to my A3000 with 150 nsec 32 bit ROMs running the minimum number of wait states). With Progressive's default MMU setup because it uses Copyback Cache mode for all RAM above 16Mb, there is the distinct possibility that ZorroIII space will not work for I/O devices. Only a properly written program which uses address translation tables will allow ZorroIII cards to work in all situations. BTW, Turn Copyback cache mode off and RAM: and RAD: Diskspeeds improve dramatically on the 040! (Either Diskspeed or the drivers for RAD: and RAM: are worst case optimized for the 040 copyback caching mechanism). -Greg – via Whap!
#16085From: Andy FinkelNov 6, 1991 10:03 AM
I think that MKS went to some trouble to make diskspeed independent of processor cache. You might want to pull out the old disk speed testing program that we used in the dawn of history (whose name escapes me at the moment). andy
#16087From: Gregory B. TibbsNov 6, 1991 12:10 PM
The A3000/030 gets 30% better results than the 040, and write through mode of the 040 is faster than copyback mode. I wasn't knocking diskspeed, I think it is a great program, just that I hadn't spent the time trying to see what is the culprit. BTW are you sure you are not talking about buffers for the device carefully not cached or actually processor caches? I guess you would want both to minimize CPU influencing the results. -Greg The 040 has additional wait states than the 030, but the fact that write through mode is faster than copyback mode baffles me: the CPU and CPU/ram speed is a constant.
#16022From: Tony PitmanNov 4, 1991 6:22 PM
thanks, tony