Login
Username:

Password:

Remember me



Lost Password?

Register now!

Sections

Who's Online
128 user(s) are online (91 user(s) are browsing Forums)

Members: 0
Guests: 128

more...

Headlines

Forum Index


Board index » All Posts (saimo)




Re: Os4.1 on the way! Yay! And help!
Quite a regular
Quite a regular


@Daedalus

Quote:
I vaguely recall something about your boot partition having to be FFS formatted - was this just for the pre-release or is it still the case?

No, also SFS can be used (it has been so for a looong while now). SFS2 and JFXS can not, though, because the SLB does not support them yet.

saimo

Go to top


Re: Good News Is Here
Quite a regular
Quite a regular


@amigakit

Good idea and good move, but, without wanting to rain on your parade, I only take it as a promising sign (here wishful-thinking)

Edit: sorry, just read the official news - well done to all!

saimo

Go to top


Re: Os4.1 on the way! Yay! And help!
Quite a regular
Quite a regular


@Skov

Quote:
4. Use Mediatoolbox to change to SFS - something I should be aware of here?

Ensure blocksize is set to 512 (mandatory).
Also, you may want to raise buffers to 1000.

saimo

Go to top


Re: AmigaOS 4.1 - software problems.
Quite a regular
Quite a regular


@abalaban

Quote:
Quote:
The problem is that links to drawers are shown as drawers, so changing those settings is of no use.

Hmm ? I'll try this evening on my own A1. Which FS are you using ?

SFS, but I think it is unlikely that it has something to do with it because, as said, ASL requesters treat links correctly - in other words, I think it is just a problem with the TextMode.

saimo

Go to top


Re: AmigaOS 4.1 - software problems.
Quite a regular
Quite a regular


This sounds strange (and serious): RAM does not get freed after usage of swap memory.
Test:
1. copy to RAM Disk more stuff than than it can hold (swapping kicks in correctly);
2. delete everything from RAM Disk;
3. free RAM is reported to be 0 (or little more) even after an "avail FLUSH".

(Problem originally reported by Menthos here.)

Edit: it does not seem to be a bug, but just a smarter way of allocating memory - here is a simple test that I did:
1. after a reboot, I opened a shell and executed avail FLUSH, which returned 185.196.544 bytes free;
2. I copied a ~28 Mb file to RAM disk;
3. I executed avail FLUSH, which returned 158.597.120 bytes free;
4. I deleted the file from RAM;
5. I executed avail FLUSH, which returned again 158.597.120 bytes free;
6. I copied the file again;
7. I executed avail FLUSH, which returned again 158.597.120 bytes free.

In a nutshell, the memory gets re-used and is re-/de-allocated only when strictly needed.

saimo


Edited by saimo on 2008/9/17 13:58:20
Go to top


Re: AmigaOS 4.1 - software problems.
Quite a regular
Quite a regular


@Cyborg

Quote:
That's not a problem of AmiDock.. every (app)docky is free in its decision if it reacts on single- or doubleclicks. AmiDock just informs the docky about what happened, but the Docky itself is responsible to react to the action.

So if one docky opens a window on single click and another docky only after doubleclick, it was the decision of the dockycoder to do it that way and not AmiDocks.

I think that there could be problem instead: a couple (or few more) times it happened to me that I clicked on the icon of an application (not a docky) and it did not start.

saimo

Go to top


Re: AmigaOS 4.1 - software problems.
Quite a regular
Quite a regular


@abalaban

Quote:
Go to the Workbench pref "text mode" tab and configure it to be underlined if you prefer...

The problem is that links to drawers are shown as drawers, so changing those settings is of no use.

saimo

Go to top


Re: Where are your AmigaOS 4.1 screenshots?
Quite a regular
Quite a regular


@AmigaPapst

Quote:
Where are your AmigaOS 4.1 screenshots?

Here's mine: just uploaded and promptly published (thanks).

saimo

Go to top


Re: AmigaOS 4.1 - software problems.
Quite a regular
Quite a regular


More not so serious issues:
- softlinks to drawers are displayed just like drawers in WB TextMode (but are correctly handled in ASL requesters);
- makelink does not trigger updating of WB listers;
- Amidock background is not 100% transparent when Icon area is set to Transparent or Transparency is set to 100% (behaviour differs);
- menus transparency is still fake (not a bug, though);
- E-UAE crashes when stack is not big enough (wasn't automatic stack enlargement implemented?)

saimo


Edited by saimo on 2008/9/17 16:08:38
Go to top


Re: AmigaOS 4.1 - software problems.
Quite a regular
Quite a regular


@Rogue

Quote:
Er... SFS is loaded with the kickstart. If one file in Kickstart cannot be loaded, then you have a much bigger problem than your SFS partitions.

There are no obscure reasons whatsoever you would be able to boot at all but SFS would not load.

Let's assume that it's just a problem with just the SYS:Kickstart/SmartFileSystem file: wouldn't the FS stored in the RDB make for a good fallback FS?
Another question: if the HD were attached to a Classic Amiga without AOS 4.x, wouldn't the FS enable the mounting of the partition (think of data recovery)?
(Yeah, I know, these are definitely extreme cases...)

saimo

Go to top


AmigaOS 4.1 benchmark
Quite a regular
Quite a regular


Out of curiosity, I performed a few tests to see how AOS 4.1 compares to AOS 4.0 (on the same machine with the same settings, of course).
The conclusion is that AOS 4.1 is (slightly) slower almost always, although an important aspect like system memory to video memory transfers seem to have improved noticeably.
Note that these tests do not prove AOS 4.1 is slower than AOS 4.0 in absolute: they cover only some aspects in a idle enviroment, so it could well be that the overall performance during normal usage is better (f.ex. thanks to the composition engine moving some of the load from the CPU to the GPU). Moreover, let's not forget that AOS 4.1 has some added functionality that may come at a cost


******************************************************************************
ramspeed -b 3

RAMspeed 
(UNIXv2.3.0 by Rhett MHollander (Alasir Enterprises), 2002-04

4Gb per pass mode

 4.0                              
4.1

 INTEGER   Copy
:      165.43 Mb/INTEGER   Copy:      164.96 Mb/s
 INTEGER   Scale
:     167.05 Mb/INTEGER   Scale:     166.50 Mb/s
 INTEGER   Add
:       168.28 Mb/INTEGER   Add:       166.23 Mb/s
 INTEGER   Triad
:     155.03 Mb/INTEGER   Triad:     153.14 Mb/s
 
------                           |
 
INTEGER   AVERAGE:   163.95 Mb/INTEGER   AVERAGE:   162.71 Mb/s



******************************************************************************
ramspeed -b 6

RAMspeed 
(UNIXv2.3.0 by Rhett MHollander (Alasir Enterprises), 2002-04

4Gb per pass mode

 4.0                              
4.1

 FL
-POINT  Copy:      225.67 Mb/FL-POINT  Copy:      224.81 Mb/s
 FL
-POINT  Scale:     225.80 Mb/FL-POINT  Scale:     224.81 Mb/s
 FL
-POINT  Add:       187.60 Mb/FL-POINT  Add:       185.34 Mb/s
 FL
-POINT  Triad:     187.60 Mb/FL-POINT  Triad:     185.23 Mb/s
 
------
 
FL-POINT  AVERAGE:   206.67 Mb/FL-POINT  AVERAGE:   205.05 Mb/s



******************************************************************************
DiskSpeed DRIVE=HD3ALL LONG
* (SFS partition)

4.0

DiskSpeed 4.3
OS4 version
Copyright 
1989-92 MKSoft Development
Copyright 
2003-04 Daniel JAndrea II St?phane Guillard
------------------------------------------------------------
CPU68020  AmigaOS Version52.8  Normal Video DMA
Device
:  HD3:    Buffers1000

Testing directory manipulation speed
.
File Create:         5603 files/sec
File Open
:          27248 files/sec
Directory Scan
:    253326 files/sec
File Delete
:         4997 files/sec

Seek
/Read:         153586 seeks/sec

Testing with a 512 byte
MEMF_FASTLONG-aligned buffer.
Create file:     13205184 bytes/sec
Write to file
:   54920384 bytes/sec
Read from file
:  79091264 bytes/sec

Testing with a 4096 byte
MEMF_FASTLONG-aligned buffer.
Create file:     24080896 bytes/sec
Write to file
:  374712832 bytes/sec
Read from file
278949888 bytes/sec

Testing with a 32768 byte
MEMF_FASTLONG-aligned buffer.
Create file:     28024832 bytes/sec
Write to file
:  194387968 bytes/sec
Read from file
233078784 bytes/sec

Testing with a 262144 byte
MEMF_FASTLONG-aligned buffer.
Create file:     27590656 bytes/sec
Write to file
:   42991616 bytes/sec
Read from file
230522880 bytes/sec


4.1

DiskSpeed 4.3
OS4 version
Copyright 
1989-92 MKSoft Development
Copyright 
2003-04 Daniel JAndrea II St?phane Guillard
------------------------------------------------------------
CPU68020  AmigaOS Version53.5  Normal Video DMA
Device
:  HD3:    Buffers1000

Testing directory manipulation speed
.
File Create:         3899 files/sec
File Open
:          25821 files/sec
Directory Scan
:    244611 files/sec
File Delete
:         4943 files/sec

Seek
/Read:         128716 seeks/sec

Testing with a 512 byte
MEMF_FASTLONG-aligned buffer.
Create file:     12938304 bytes/sec
Write to file
:   43209233 bytes/sec
Read from file
:  70056576 bytes/sec

Testing with a 4096 byte
MEMF_FASTLONG-aligned buffer.
Create file:     23139328 bytes/sec
Write to file
:  250321408 bytes/sec
Read from file
238501888 bytes/sec

Testing with a 32768 byte
MEMF_FASTLONG-aligned buffer.
Create file:     26972160 bytes/sec
Write to file
:  103411712 bytes/sec
Read from file
174030848 bytes/sec

Testing with a 262144 byte
MEMF_FASTLONG-aligned buffer.
Create file:      5570560 bytes/sec
Write to file
:    5832704 bytes/sec
Read from file
:   5734400 bytes/sec

(side noteI don't have the results anymore nor a partition to repeat the test, but JXFS turned out to be slightly slower).



*******************************************************************************
* Custom program that renders stuff in system memory and then transfers it to
* video memory for showing in a window on Workbench.

* 4.0

tiles rendering speed: 461.095101 fps, 2.168750 ms/loop
bobs rendering speed: 288.548242 fps, 3.465625 ms/loop
video data transfer speed: 107.346528 fps, 9.315625 ms/loop
total speed: 66.861680 fps


* 4.1 - composition ON

tiles rendering speed: 468.864469 fps, 2.132812 ms/loop
bobs rendering speed: 259.109312 fps, 3.859375 ms/loop
video data transfer speed: 131.497843 fps, 7.604688 ms/loop
total speed: 73.520965 fps


* 4.1 - composition OFF

tiles rendering speed: 468.864469 fps, 2.132812 ms/loop
bobs rendering speed: 277.898393 fps, 3.598438 ms/loop
video data transfer speed: 116.257947 fps, 8.601562 ms/loop
total speed: 69.754768 fps



******************************************************************************
* A script opens and closes 2 windows 1000 times.

* 4.0 /  4.1 - composition ON / 4.1 - composition OFF

seconds elapsed: 31 / 32 / 30




saimo

Go to top


Re: AmigaOS 4.1 - software problems.
Quite a regular
Quite a regular


@xenic

Quote:
That should have been user configurable in ASL prefs.

Glad to hear it
Indeed, I really hoped it was so

Quote:
Maybe it will be added in an update.

Can't wait!

saimo

Go to top


Re: AmigaOS 4.1 - software problems.
Quite a regular
Quite a regular


@Rigo

Quote:
Why would you need a 68K version of SFS on an A1?

Because that's the only one that can be installed in the RDB - just in case for some obscure reason the filesystem cannot be loaded from file.

saimo

Go to top


Re: AmigaOS 4.1 - software problems.
Quite a regular
Quite a regular


These are not really problems, but here they come anyway...

- interpretation/handling of BUFSIZE tooltype of AsyncWB seem to have changed: I used to have it set to 1 Mb, but that now makes file copying crawl (maybe the magnitude has changed so that it caused VM to kick in?) - I eventually found out that 32768 is better than leaving it unspecified and than 65536;

- a 68k version of the latest SmartFilesystem (1.282) is not included, which means that the installed SmartFilesystem cannot be updated (the last public one is version 1.279);

- (as mentioned in here on AW.net) ASL now requires a double-click to enter directories.

saimo


Edited by saimo on 2008/9/15 18:36:58
Go to top


Re: Thank you, Dennis Smith!
Quite a regular
Quite a regular


@Chris

Great! Thanks a lot!

saimo

Go to top


Re: Thank you, Dennis Smith!
Quite a regular
Quite a regular


@Chris

The problem is: I have no idea of who Dennis Smith is... I do not even know whether he is registered here!
I'm just hoping this message somehow reached him...

saimo

Go to top


Thank you, Dennis Smith!
Quite a regular
Quite a regular


Hi Dennis,

I just wanted to say "thank you" (you know for what).
I tried sending you an email, but it was rejected with this message: "Diagnostic-Code: smtp;550 5.1.1 Not our Customer".

saimo

P.S. apologies for using amigans.net for such a private reason, but I have no other means to try to reach him.

Go to top


Re: Bl***y Amiga OS4. :-)
Quite a regular
Quite a regular


@joerg

Thanks a lot for the clarification

saimo

Go to top


Re: Bl***y Amiga OS4. :-)
Quite a regular
Quite a regular


@Swoop

Quote:

Swoop wrote:
@saimo

The thing is I had been sticking with FFS, just in case my A1 blew up and I had to transfer the hard drive over to my A1200.

SFS also comes as M68k binaries that can be installed in the RDB*, thus - I guess - access to SFS partitions should be possible also on AOS 3.x (f.ex., see this post by Rogue). Since it is such a delicate issue, it would be nice if anybody could give a confirmation.

*Indeed, as Joerg explained, those are the only binaries allowed in the RDB.

saimo

Go to top


Re: Bl***y Amiga OS4. :-)
Quite a regular
Quite a regular


@Swoop

Quote:
Well would you believe it , that OS4 again.
I managed to screw up one of my partitions, Status validating, nothing I could do to get it to change.

Quick suggestion to forget about validation issues altogether: use SFS

saimo

Go to top



TopTop
« 1 ... 27 28 29 (30) 31 32 33 »




Powered by XOOPS 2.0 © 2001-2023 The XOOPS Project