FreeArc

[JangoFatXL]

sorry, I did not see it, you already mentioned that point for v2.0

[Bulat Ziganshin]

new version:

• Multithreaded deflate compressor - fastest on the planet!
• -mt option now passed down to 7z.dll
• new translation: belarusian
• ArcShellExt: new "Add to .zip" and ""Add to .7z" commands
• GUI: added 10-seconds delay before deleting temporary file

.zip compression test on 800 mb dataset:

zip: 68 secs
PowerArchiver, 7-zip: 31 secs
FreeArc: 21 secs

"running" pictures directly from archive always opens (and it always did this way before too) default Image Viewer saying "Image cannot be displayed"

can anyone confirm that now it works ok?

[Black_Fox]

Multithreaded deflate compressor - fastest on the planet!

That is very cool, another win for FA!

* GUI: added 10-seconds delay before deleting temporary file
can anyone confirm that now it works ok?

Yes, it's now fixed, thanks a lot

Could the window be at least resizable at this point, if not wider?

i don't found a way to do it with NSIS Modern UI

How about displaying the selection box and the description box in two rows then, as opposed to current two columns? Can that be done?

[JangoFatXL]

ArcShellExt: new "Add to .zip" and ""Add to .7z" commands

check boxes in settings menu where you can choose whether you want that two options to be shown or not (like in 7z/winrar) would be nice

edit: forget it, there is already that option :P

[Vacon]

Hello everyone,

new version:

• Multithreaded deflate compressor - fastest on the planet!
• -mt option now passed down to 7z.dll
• new translation: belarusian
• ArcShellExt: new "Add to .zip" and ""Add to .7z" commands
• GUI: added 10-seconds delay before deleting temporary file

.zip compression test on 800 mb dataset:

zip: 68 secs
PowerArchiver, 7-zip: 31 secs
FreeArc: 21 secs

can anyone confirm that now it works ok?

I'm not using Window's default viewer but XNView-portable. Works like a charme from within arc-, zip- and 7z-files, as well as from flat archives as from within folders inside archives on XP.

Best regards!

[Vacon]

Hello everyone,

made a quick test regarding new zip-feature (i386-folder containing 2606 files (518.963.955 Bytes), nested in two folders, created by XP during installation of SP3 on a Athlon XP 1700+ with 1,25 GB RAM):
7-Zip -> 6min 38sec resulting in 253.478.776 Bytes
FreeArc -> 3min 09sec resulting in 258.441.237 Bytes at ~2,600kB/s

Best regards!

[Bulat Ziganshin]

Vacon, your cpu isn't multicore/HT, so you just measured the difference of 7zip versus zlib deflate

[IsName]

Vacon, your cpu isn't multicore/HT, so you just measured the difference of 7zip versus zlib deflate

So you used zlib deflate, and not your own one? Maybe you can use some code from Tornado.

[Vacon]

Hello everyone,

Vacon, your cpu isn't multicore/HT, so you just measured the difference of 7zip versus zlib deflate

I do know that
but I never had a real reason, nor the opportunity to compare these really (thanks to the dialog FA gives during compression).

Best regards!

[Bulat Ziganshin]

Maybe you can use some code from Tornado.

it will not make much difference

How about displaying the selection box and the description box in two rows then, as opposed to current two columns? Can that be done?

look at alternative version at http://freearc.org/download/FreeArc-...lpha-win32.exe

i can't do anything else, there are just 2 fixed layouts. eventually i will start using InnoSetup

[Black_Fox]

That looks much better at first sight, but now some of the descriptions are cropped Let's keep the first one (with two columns) then, that can be at least scrolled

[Bulat Ziganshin]

yes, it's why i use it

[tonifelfe]

Hi Bulat, i have a 9GB (!) .arc and cannot unpack it:
ERROR: can't allocate memory required for (de)compression in 4x4:tor:16mb:c3
What can i do - impossible decompression is not a value!
Thanks in advance for your help.
Toni

[Bulat Ziganshin]

probably you are using unarc; use arc.exe/freearc.exe for decompression

[tonifelfe]

I have used FreeArc GUI. Trying now with "Arc.exe x -r name.arc"
Is there any clue to limit resource requests while packing/unpacking?
Using 100% RAM or using 100% CPU should be an option, not default.

[Bulat Ziganshin]

try with -mt-. and remove -r

Using 100% RAM or using 100% CPU should be an option, not default.

and what should be the default?

[tonifelfe]

Thanks. Difficult question. But: action has to run with low resources too - time is not crucial.
If i may dream: selection between 50%/100% cpu/ram would be fine.
The point is, that unpacking has to work at almoust every machine, when i deliver arc.

[JangoFatXL]

Why dont you use asymetric compression? Check the box "Fast, low-memory decompression" in compression tab. If you want max possible asymetric compression use -m9x. If you want to ensure it can be unpacked on nearly every machine add -ld128m

[tonifelfe]

Thanks, but it IS compressed now, i believe without -m9x and -ld128m.
Got this error using ARC.exe:

arc.exe x -mt- alteserver
FreeArc 0.67 (August 12 2010) extracting archive: alteserver.arc
35,407 recovery sectors (69 mbytes) present
Scanning archive for damages...
Archive integrity OK
Extracting 286,934 files, 21,778,540,831 bytes. Processed 96.4%
ERROR: can't allocate memory required for (de)compression in 4x4:lzma:16mb:h4mb:
normal: 96.4%

Can i unpack the whole arc???

[JangoFatXL]

While we're on it, just tested -ld128m, -ld96m and -ld64m by myself. I noticed that FreeArc always needs about 30mb more for extraction than the switch connotes (just looked into the task manager).

Is that normal? Can the archive be extracted on a machine with f.e. 128m if the -ld128m (in combination with -m9x) switch needs about 166mb? (I cannot test it in a low-memory environment).

[Bulat Ziganshin]

Is that normal? Can the archive be extracted on a machine with f.e. 128m if the -ld128m (in combination with -m9x) switch needs about 166mb?

1. -md limits memory used by extraction algorithm, not the program at whole
2. when memory allows, all algorithms are decoded simultaneously, when memory is low, decompression goes through the temporary disk files
3. i recommend to set -ld to 3/4 of RAM on the worst computer used for decompression. for example my Max/Ultra settings are configured to ld800/1600

Thanks. Difficult question. But: action has to run with low resources too - time is not crucial.
If i may dream: selection between 50%/100% cpu/ram would be fine.

freearc is able to utilize 100% of your CPU free time. unlike other compressors, it lowers priority of compression threads, so it doesn't affect your other activity

The point is, that unpacking has to work at almoust every machine, when i deliver arc.

for this, don't use max/ultra settings. the rest will decompress on any computer with 256 mb of RAM

Can i unpack the whole arc???

heh, try to use unarc. other option is to extract files in several steps, say all files less than 1mb first, then 1 to 10 mb...

[JangoFatXL]

2. when memory allows, all algorithms are decoded simultaneously, when memory is low, decompression goes through the temporary disk files

Well, now you mention it, I already saw that behaviour when I created an archive with ultra settings under Win7 64bit and later tried to extract with WinXP 32bit. It worked, but needed much more time for extraction.

[JangoFatXL]

- updated german lang file
- added some new tooltips

[Bulat Ziganshin]

thanks, Jango

[zyzzle]

It is wonderful news about the multithreaded deflate decompressor, but I seem unable to use it! Please forgive this stupid question, but when I try to decompress a ZIP file with commandline arc.exe, it says not a valid archive, and does not decompress the file. Is it possible to directly decompress .ZIP files wih commandline ARC.EXE? What are the options? Thanks.

[Bulat Ziganshin]

- added some new tooltips

please provide english versions of your tooltips:

--Main
1166 Logfile:=Hier k?nnen Sie eine logdatei definieren
1447 Temporary directory:=Hier k?nnen Sie ein Verzeichnis f?r tempor?re Dateien definieren. W?hlen Sie einen Ort auf der schnellsten Festplatte mit hoher Kapazit?t.

--Interface
1068 Language:=Programmsprache ?ndern\n(Programmneustart zum ?bernehmen der Spracheinstellungen erforderlich)
1361 Add captions to toolbar buttons=Wenn Sie diese Option aktivieren, wird im Hauptfenster unter den Symbolen zus?tzlich eine kurze Beschreibung eingeblendet\n(Programmneustart zum ?bernehmen der Einstellungen erforderlich)

--Explorer integration
1172 Associate FreeArc with .arc files=Aktivieren Sie diese Option um FreeArc mit .arc- Archiven zu verkn?pfen (empfohlen)
1471 Associate FreeArc with other archives=Aktivieren Sie diese Option um fremde Archive mit FreeArc zu verkn?pfen
1423 Make it cascaded=Aktivieren Sie diese Option um die folgenden Befehle in einem Untermen? anzuzeigen (empfohlen)

[JangoFatXL]

--Main
1166 Logfile:=Here you can define a logfile
1447 Temporary directory:=Here you can define a directory for temporary files. Choose the fastest drive with high capacity.

--Interface
1068 Language:=Change program language\n(Program restart needed to take effect changes)
1361 Add captions to toolbar buttons=If you activate this option, short descriptions will be shown under the symbols in main window\n(Program restart needed to take effect changes)

--Explorer integration
1172 Associate FreeArc with .arc files=Activate this option to open .arc- archives with FreeArc (recommended)
1471 Associate FreeArc with other archives=Activate this option to open other archives with FreeArc
1423 Make it cascaded=Activate this option to arrange the following commands in a submenu (recommended)

feel free to improve english language in this tooltips

[Black_Fox]

1166, 1447: define -> specify, also the word order should be changed to something like "logfile can be specified here", but that still doesn't sound completely correct...
1068, 1361: (Program needs to be restarted for the changes to take effect)

[d33j4y]

Some programs like WinRAR, 7z, FreeArc use multi-thread compression to increase speed...

But none at the moment use GPU. Is FreeArc planned to use GPU programming?

[Bulat Ziganshin]

no, it's impossible