Overview Features Coding ApolloOS Performance Forum Downloads Products Order Contact

Welcome to the Apollo Forum

This forum is for people interested in the APOLLO CPU.
Please read the forum usage manual.
Please visit our Apollo-Discord Server for support.



All TopicsNewsPerformanceGamesDemosApolloVampireAROSWorkbenchATARIReleases

GOLD 2.10 Releasepage  1 2 3 

Manfred Bergmann

Posts 226
24 Jul 2018 07:33


David Wright wrote:

everything in software updates to rounding off my tax payment means the second digit after the decimal effects the first digit after the decimal.
  2.8, 2.9. 2.95 then 3. or whatever.
  You coders do it different?
 
  Make a joke about it but the sequencing doesn't make sense. Just saying, never seen this before.

The versioning depends a bit on what is new, or has changed in the software.
You don't increment the major number just because the minor number is at 9.
You increment it if it contains major new and possibly API breaking changes.
And if you are at 9, and it's just a bug-fix release, or only contains minor new enhancements, then it is a 10.


Marlon Beijer

Posts 182
24 Jul 2018 11:26


David Wright wrote:

everything in software updates to rounding off my tax payment means the second digit after the decimal effects the first digit after the decimal.
  2.8, 2.9. 2.95 then 3. or whatever.
  You coders do it different?
 
  Make a joke about it but the sequencing doesn't make sense. Just saying, never seen this before.

Software/hardware versioning is not math, you don't use the decimal like in math.

Maybe a bit tl;dr but, here's an explanation:
EXTERNAL LINK 
SDL went all the way up to 1.2.15 before switching to 2.0 because the changes were so big it broke compatibility with applications using SDL-1.2.

Major.Minor.Patch/"Release candidate"


David Wright

Posts 373
24 Jul 2018 14:21


Thanks Tim, Marlon and others. Interesting to learn, though not terribly interesting. I was looking at it from a math/decimal position.

Maybe Gunnar can link me another Sesame Street lesson instead of the explanatory one Marlon did.:)

At any rate, the really interesting number here is three, drei, trois, tres. Really soon I bet.


Gunnar von Boehn
(Apollo Team Member)
Posts 6197
24 Jul 2018 16:35


David Wright wrote:

  Maybe Gunnar can link me another Sesame Street lesson instead of the explanatory one Marlon did.:)

 
No problem. We are here to help. ;)
 
Our GOLD 2.xx number range is for Cores without AGA.
 
GOLD 3... will introduce major changes.
Therefore the "major-number" the first number will change.
 
 
Everytime you introduce very major changes like incompatibilities you change the major number.
 
 
Lets look at Amiga OS for numbering as example.
 
AMIGA OS 3.1 was based on OS 3.0 with some fixes / improvements
AMIGA OS 3.5 was based on OS 3.1
And based on this was OS 3.9
 
Now AMIGA OS 4.0 is based on OS 3.1.
So it did NOT include what is in OS 3.9
The major number was changed from 3 to 4 because OS 4.0 is only for PPC - means of course some incompatibility.
 
So even when OS 4.0 sound mathematically "more" feature wise
in reality it started with features below "3.9".
Of course they might have added / reinvented these features meanwhile.
 
If now someone would continue to develop OS 3.X base on the last 3.9 version than a label "3.10" or "3.11" would be correct as long it stays compatible.
 
And feature wise such "OS 3.11" release could be well better or more than "OS 4.0"


Vojin Vidanovic
(Needs Verification)
Posts 1916/ 1
24 Jul 2018 19:31


Gunnar von Boehn wrote:

  So even when OS 4.0 sound mathematically "more" feature wise
  in reality it started with features below "3.9".
  Of course they might have added / reinvented these features meanwhile.
 
 
  And feature wise such "OS 3.11" release could be well better or more than "OS 4.0"

On OS 3.9 - No they havent implemented it all, and OS 4.1 did not carry all OS 4.0 feats (e.g. Warp3D driver for all cards, WarpOS compatibility,OS 4.0 look as default).  OS 4.1 did implement Dockie, EnhancerOS Dockie X and now with EnhancerOS 2.0 they promote even newer eyecandies.

Well OS 3.11 is by number and in true just OS 3.1 fixed, so no, its no better or more then 4.0. Most of things that will be fixed there are fixed by now. Its juts improvement for vanilla OS 3.1 users.

I do have OS 3.5 and will get OS 3.9, so nothing interesting for I there.


Gunnar von Boehn
(Apollo Team Member)
Posts 6197
24 Jul 2018 19:35


Vojin Vidanovic wrote:

Well OS 3.11 is by number and in true just OS 3.1 fixed,

You mean 3.1.4 ?
The new fix release of 3.1 ?
Yes as the numbering also suggest is 3.1 + some fixes.
It has less features than 3.9


Vojin Vidanovic
(Needs Verification)
Posts 1916/ 1
24 Jul 2018 22:55


Gunnar von Boehn wrote:

  Yes as the numbering also suggest is 3.1 + some fixes.
  It has less features than 3.9

It will have far less feats and fixes then my OS 3.5 CD, I believe. But yet is good work and nice thing to exist.

There are 3.1 purists, and 3.1.4 is nice for 00 Miggys and fast emulation. So with bit 080 code even Vamps could benefit from most lightweight modern AmigaOS (that can be expanded in users ways) :-)


Leigh Russ

Posts 151
25 Jul 2018 00:44


Knight Stone wrote:

it's not gonna happen, not at least until you use the EIDE channel, that's on the V4. The IDE port on the A600 was never really meant to go above a few MB's a second, and as such, is probably pretty near maximum throughput on an accelerated A600 anyway. The only way to get faster, is to use an EIDE addin board, and maybe a mechanical PIO mode 4/5 drive, but even then, the gains will be minimal.
 
  Wait for the V4, that's the only way to get 20MB's on the A600

I guess you are quite recent to the vampire scene as quite some time ago there was a beta core that done exactly this and allowed you to tweak the A600 IDE speed so it has already happened and is possible. And we are fully aware it will not reach the speeds of the A500 Vampire, but even an extra 512k-1mb/s is a great improvement on the standard 3mb/s ish.

I understand why it isn't enabled, however I was only asking as both myself and several others would of liked to try this feature, understanding the risks involved. Just seems a shame to not leave it up to the users to decide if to try it.

I guess we can only hope for the ability to boot from SD card but it seems to have gone all quiet on that front for quite some time.

Nice to see regular updates nonetheless. Cheers


Leigh Russ

Posts 151
25 Jul 2018 00:45


Renaud Schweingruber wrote:

As Gunnar said regarding A600 IDE, stability of FastIDE on A600 wasn't yet good enough for a mass release.
 
  If you want to squeeze some more performance on A600, you can give a try to this tool written by nonarkitten. I got nearly +10% improvement depending on configuration.
  EXTERNAL LINK 

Thanks Renaud, I will check it out. Nothing lost if it doesn't do much afterall


Adam Whittaker
(Needs Verification)
Posts 270/ 1
25 Jul 2018 13:03


I agree Leigh Russ - why not give us the option to try? we have asked about this at least 10 times now and we just get looked over!!! why? come on somebody tell us why please?


Peter Heginbotham

Posts 214
25 Jul 2018 13:44


I had the beta core and was seeing 4-6 MB but it was flaky and would lock the computer when copying between disks.


Adam Whittaker
(Needs Verification)
Posts 270/ 1
25 Jul 2018 13:58


yes we have been told this so give us the speed setting options so we can experiement as not all amiga 600's are the same?


Gunnar von Boehn
(Apollo Team Member)
Posts 6197
25 Jul 2018 14:36


Adam Whittaker wrote:

yes we have been told this so give us the speed setting options so we can experiement as not all amiga 600's are the same?

Yes you are fully correct.
There is big different in speed you can get out of those 30 year old A600s. I have seen anything from 4MB/sec to 9 MB/sec.

The main problem that I saw is that not only different A600 reach different speed - but also the performance varies on the same A600 with different temperature / runtime.

We saw users using the Beta cores and finding good performance values which run stable for a long time but after some hours became unstable. This means to be sure users will need to test in depth for hours or days. I see here a risk for users not testing carefully enough and then loosing data.


Adam Whittaker
(Needs Verification)
Posts 270/ 1
25 Jul 2018 17:10


to be fair i am pretty sure we all keep a backup of our amiga drive these days so loosing data shouldnt be a big problem - if this is a problem for some users they simply just dont have to use the the speed boost settings.


Matthias Overloeper

Posts 3
03 Aug 2018 17:45


Hi,
it seems that there is something wrong with your files.

First I had downloaded (today) the EXE... but on my Amiga I can't unpack it (tested with OS3.9 Unarchiver AND "lha x ..." in shell
Than I had downloaded both JIC-Files but while flashing I' get an error.

Now I have flashed my V600 with a backup of 2.9 core. THIS works fine.
In the past I have had never problems like that.


Renaud Schweingruber
(Apollo Team Member)
Posts 378
04 Aug 2018 16:24


OS3.9 Unarchiver is well known to sometimes fail to unpack files. I use LHA when it happens.


Peter Heginbotham

Posts 214
04 Aug 2018 17:26


Is there going to be a X10 release?


Knight Stone
(Needs Verification)
Posts 136/ 1
05 Aug 2018 01:02


Leigh Russ wrote:

  I guess you are quite recent to the vampire scene as quite some time ago there was a beta core that done exactly this and allowed you to tweak the A600 IDE speed so it has already happened and is possible. And we are fully aware it will not reach the speeds of the A500 Vampire, but even an extra 512k-1mb/s is a great improvement on the standard 3mb/s ish.

No, i am not "quite new to the Vampire scene", Whatever the f**k the "Vampire scene" is. what a ridiculous statement to make, and who the f**k are you, some sort of Oracle? I lurk here, and don't post much, because 90% of the posts on here, are sycophantic nonsense, begging for GOLD3.0, and i remember the core you mention, it was for Beta testers only, and wasn't implemented, as it wasn't that reliable.

It's been stated we wont have RELIABLE, and make no mistake RELIABLE is the key word here, high speed IDE, until the V4, Gunnar has said so on occasion. And please, don't make any assumptions about me, or what you think i do, or don't know.

thanks


ExiE CZEX

Posts 48
05 Aug 2018 14:36


Vojin Vidanovic wrote:

Gunnar von Boehn wrote:

  Yes as the numbering also suggest is 3.1 + some fixes.
  It has less features than 3.9
 

 
  It will have far less feats and fixes then my OS 3.5 CD, I believe. But yet is good work and nice thing to exist.
 
  There are 3.1 purists, and 3.1.4 is nice for 00 Miggys and fast emulation. So with bit 080 code even Vamps could benefit from most lightweight modern AmigaOS (that can be expanded in users ways) :-)

That is incorrect statement. You are right OS 3.1.4 can't be compared to 3.5/3.9 but it will bring some fixes that are not available even in 3.9BB2. And probably usable as replacement of some OS3.9 parts...



Vojin Vidanovic
(Needs Verification)
Posts 1916/ 1
05 Aug 2018 14:37


ExiE CZEX wrote:

  That is incorrect statement. You are right OS 3.1.4 can't be compared to 3.5/3.9 but it will bring some fixes that are not available even in 3.9BB2. And probably usable as replacement of some OS3.9 parts...

Well, then I would buy such separate "BB3" install as OS 3.9 user


posts 50page  1 2 3