Puppy Users Group ---The PUG ....
..and if we standardise building with a zdrv it becomes even easier..
-replace zdrv
-fix initrd
-replace vmlinuz
-overwrite modules.order and some small stuff in iso and it's nearly doable in a running puppy
.. ok I'll work toward that with hopefully a demo in next spup.
-replace zdrv
-fix initrd
-replace vmlinuz
-overwrite modules.order and some small stuff in iso and it's nearly doable in a running puppy
.. ok I'll work toward that with hopefully a demo in next spup.
Puppy Linux Blog - contact me for access
- Lobster
- Official Crustacean
- Posts: 15522
- Joined: Wed 04 May 2005, 06:06
- Location: Paradox Realm
- Contact:
p310don idea makes a lot of sense.
A standard Puppy 5.3
for those on dial up.
Also release a large version for customisation.
Puppys can then customise and create ISO with different options.
Many possibilities to offer for release
to kennel members on slower connection . . .
roll on Spup Alpha4
Puppy Linux
The Awesome Power of Penguins
A standard Puppy 5.3
for those on dial up.
Also release a large version for customisation.
Puppys can then customise and create ISO with different options.
Many possibilities to offer for release
to kennel members on slower connection . . .
roll on Spup Alpha4
Puppy Linux
The Awesome Power of Penguins
Shouldn't leave
@James C, you shouldn't be leaving. All help and input is important so that the builders can get it right. Your input is important.
Please check ICE threads. for my reports which were done on 8-9 yearold PCs as well as 2-3 year old 64bit PCs. PAE performs extremely well. If you have some pre-2007 PC and are having trouble, be rest assure for all I know about the hardware engineering spec, and the Intel internal reports, PAE runs performance gains over non-PAE operations for memory frame references in the 32bit arena. This was even before 64bit PCs came onto the customer landscape.
Not to offend, but Barry builds, maintains, and provides WARY, He has directily and indirectly indicated the platforms he is serving. That is small RAM platforms. With what Bary does for Bones, Woof, Quirky, WARY and forum work, I say he has his hands full bringing those release to us
All we have to do is cover the rest of the PCs of the world thru the packaging efforts for user productiive proposes.
The OP, on this thread, asks about Puppy. This, to me, means the class of PCs that has been referenced. If we are still talking about 2007+ PCs, then I cannot see anything wrong with provisioning a PUP for that covers from 512MB all the way up to 64GB, the architecture of 32bit machines. When IBM completed version 4 of OS2, it was built (upgraded) to run on 32bit machines all the way to 64GB. Does anyone remember how old IBM's OS2 is?
We won't be hurting the community or harming the product by doing so. So far, we have seen demonstrations of upsides to this improvement for that community of systems. From my ponit of view, this could have a tremendous plus if a little iLinux distro like Puppy addresses and outflanks its competitor in both system awareness and performance...Wow!
I had mentioned before that future PAE discussion should be on that thread. Maybe if someone feels the discussion needs to be re-addressed then a thread should be opened or re-opened. As fas as I know, excepting some very small increases in kernel size, its benefits to us and to users is more than significant.
But please don't leave because we are discussing including PAE. Has anyone seen the recent industry announcements? Our PC industry is seeing tital wave growth. We are discussing a OS that is going to be riding on that tilal wave.
If anyone feels strongly, please open a thread and allow us to address concerns and ideas you may have on PAE.
I for one will not leave this discussion or thread because PAE isn't included. Our mission is to put on the table things which would make Puppy meet today's and near-future users system facilities.
Further James C; you and I are the only Live media production users in the total Puppyland. Eveyone else installs it on HDD/USBs.
Edited: ALL Hardinfo reports I have ever produced are on Live media systems with a SWAP partition on the HDD/USBs. This may be why I have not had issues with any PUPs that i run. Linux architecture suggest that I should expect greater latitude by providing SWAP. PUPs, at boot time, detects it and picks it up; same as all other distro i test. Could that be cause for some performance degradation you saw on your small RAM system.
Please check ICE threads. for my reports which were done on 8-9 yearold PCs as well as 2-3 year old 64bit PCs. PAE performs extremely well. If you have some pre-2007 PC and are having trouble, be rest assure for all I know about the hardware engineering spec, and the Intel internal reports, PAE runs performance gains over non-PAE operations for memory frame references in the 32bit arena. This was even before 64bit PCs came onto the customer landscape.
Not to offend, but Barry builds, maintains, and provides WARY, He has directily and indirectly indicated the platforms he is serving. That is small RAM platforms. With what Bary does for Bones, Woof, Quirky, WARY and forum work, I say he has his hands full bringing those release to us
All we have to do is cover the rest of the PCs of the world thru the packaging efforts for user productiive proposes.
The OP, on this thread, asks about Puppy. This, to me, means the class of PCs that has been referenced. If we are still talking about 2007+ PCs, then I cannot see anything wrong with provisioning a PUP for that covers from 512MB all the way up to 64GB, the architecture of 32bit machines. When IBM completed version 4 of OS2, it was built (upgraded) to run on 32bit machines all the way to 64GB. Does anyone remember how old IBM's OS2 is?
We won't be hurting the community or harming the product by doing so. So far, we have seen demonstrations of upsides to this improvement for that community of systems. From my ponit of view, this could have a tremendous plus if a little iLinux distro like Puppy addresses and outflanks its competitor in both system awareness and performance...Wow!
I had mentioned before that future PAE discussion should be on that thread. Maybe if someone feels the discussion needs to be re-addressed then a thread should be opened or re-opened. As fas as I know, excepting some very small increases in kernel size, its benefits to us and to users is more than significant.
But please don't leave because we are discussing including PAE. Has anyone seen the recent industry announcements? Our PC industry is seeing tital wave growth. We are discussing a OS that is going to be riding on that tilal wave.
If anyone feels strongly, please open a thread and allow us to address concerns and ideas you may have on PAE.
I for one will not leave this discussion or thread because PAE isn't included. Our mission is to put on the table things which would make Puppy meet today's and near-future users system facilities.
Further James C; you and I are the only Live media production users in the total Puppyland. Eveyone else installs it on HDD/USBs.
Edited: ALL Hardinfo reports I have ever produced are on Live media systems with a SWAP partition on the HDD/USBs. This may be why I have not had issues with any PUPs that i run. Linux architecture suggest that I should expect greater latitude by providing SWAP. PUPs, at boot time, detects it and picks it up; same as all other distro i test. Could that be cause for some performance degradation you saw on your small RAM system.
I've been reading more on kernels and stuff the evening, and doing my own thinking on what direction I would like to see puppy to go in.
I'm sticking with my suggestion of something that is user selectable from the cd / ISO.
For me, using Linux is about choice.
James C isn't here at the moment, but I have spoken to him since the earlier posts. He can correct me if I am wrong here, but his issue with PAE is that it doesn't work on ALL systems, and as such shouldn't be included in a default puppy for live cd installation. This absolutely makes sense, and I agree with him on that matter.
If 01micko can do what he says, and build on jrb's method for switching kernels, something I think also iguleder has been working on, then that could be incorporated into the installation idea I proposed. (if that idea is do-able)
To build on my idea, the base live cd would have puppy as it is now, with a standard kernel and everything that goes into making puppy. Through the installation and customisation process, the end user could select what additions they would want, including kernel features. If that is indeed do-able, there is much more scope beyond PAE which we've mentioned a bit too much already. I am definitely not an expert at anything kernel, but I believe there is specific kernel support for atom processors, low latency kernels for media production, older machines etc.
If a user can tailor their puppy to suit their usage and their machine, it could make for a more efficient, and hopefully faster and better puppy experience.
I'm sticking with my suggestion of something that is user selectable from the cd / ISO.
For me, using Linux is about choice.
James C isn't here at the moment, but I have spoken to him since the earlier posts. He can correct me if I am wrong here, but his issue with PAE is that it doesn't work on ALL systems, and as such shouldn't be included in a default puppy for live cd installation. This absolutely makes sense, and I agree with him on that matter.
If 01micko can do what he says, and build on jrb's method for switching kernels, something I think also iguleder has been working on, then that could be incorporated into the installation idea I proposed. (if that idea is do-able)
To build on my idea, the base live cd would have puppy as it is now, with a standard kernel and everything that goes into making puppy. Through the installation and customisation process, the end user could select what additions they would want, including kernel features. If that is indeed do-able, there is much more scope beyond PAE which we've mentioned a bit too much already. I am definitely not an expert at anything kernel, but I believe there is specific kernel support for atom processors, low latency kernels for media production, older machines etc.
If a user can tailor their puppy to suit their usage and their machine, it could make for a more efficient, and hopefully faster and better puppy experience.
Hmm - do you mean a bit of a petshop perhaps, where a new user might waltz in to a Puppy website, select from a menu of RAM and CPU (plus 32bit and 64bit, Intel vs AMD etc) hardware limitations, perhaps choose a WM (JWM, Icewm, openbox, lxde, compriz, E17 whatever) and have a SUGGESTED skeleton tailor-made for their hardware to download, over which they simply add their preferred apps with PPM? And just upgrade the kernel as they upgrade their machine? OOOOooooh I like it!
Who was it who had in his signature "Puppy Linux: Have it your way"?
[EDIT: sorry p310don - you practically stated the above earlier here http://www.murga-linux.com/puppy/viewto ... 226#531226. Bedtime for me]
Who was it who had in his signature "Puppy Linux: Have it your way"?
[EDIT: sorry p310don - you practically stated the above earlier here http://www.murga-linux.com/puppy/viewto ... 226#531226. Bedtime for me]
Search engines for Puppy
[url]http://puppylinux.us/psearch.html[/url]; [url=https://cse.google.com/cse?cx=015995643981050743583%3Aabvzbibgzxo&q=#gsc.tab=0]Google Custom Search[/url]; [url]http://wellminded.net63.net/[/url] others TBA...
[url]http://puppylinux.us/psearch.html[/url]; [url=https://cse.google.com/cse?cx=015995643981050743583%3Aabvzbibgzxo&q=#gsc.tab=0]Google Custom Search[/url]; [url]http://wellminded.net63.net/[/url] others TBA...
A petshop sounds like it would be a great thing for someone who knows the details of the hardware he or she has and is already familiar enough with the application programs available to know which ones he or she might want.
What about the newcomer to Puppy? Or the person who doesn't want to become an expert on his or her computer, or Linux in general? Are we chopped liver?
What about the newcomer to Puppy? Or the person who doesn't want to become an expert on his or her computer, or Linux in general? Are we chopped liver?
'standard/basic install' option with 'custom install' option?
ie automated or user defined, one click for noobs, a few more for the more experienced/experimental?
a user may experiment more - non destructively - with experience, using pupsave files? not sure i'm making sense anymore, time for bed...
*edit* but before i do - one more thought, and i'm thinking out of the top of my hat here, not really knowing what i'm on about... but would a kind of online remastering service be possible via repositories? boot to basic puppy and use as is or select options to be (auto?) downloaded followed by prompt to remaster?
night night
ie automated or user defined, one click for noobs, a few more for the more experienced/experimental?
a user may experiment more - non destructively - with experience, using pupsave files? not sure i'm making sense anymore, time for bed...
*edit* but before i do - one more thought, and i'm thinking out of the top of my hat here, not really knowing what i'm on about... but would a kind of online remastering service be possible via repositories? boot to basic puppy and use as is or select options to be (auto?) downloaded followed by prompt to remaster?
night night
ohm's where the art is
G'day Flash -Flash wrote:...
What about the newcomer to Puppy? Or the person who doesn't want to become an expert on his or her computer, or Linux in general? Are we chopped liver?
I think I see where you're coming from - you know that chopped liver is seen as a great treat for puppians - a bit like those pigs ears - and pigs ears come into the Puppy Process and habitually get turned into silk purses...
Ok so we have a diverse market that shares one crucial trait - they suspect strongly that they have been sold a Dog, or a steaming pile of barker's nests, from the Big End of town. Too much in-breeding with the OS that was dictated to them when they bought their computer - and while some might want to break away from that crowd completely and dance with wolves, others prefer just the occasional stroll with an unusual little "bitzer" (mongrel) they picked up as a stray - but continually amazes it's new owner with its abilities, loyalty, sense of fun, and downright doggedness in work ethic.
Much has been said elsewhere on the forums about whether we should dress up our "lamb" as XP, Vista or Win7 mutton without showing the wolf it actually can be, when needed, by its Owner. (Yay to the Hungarian Puppy Team in their XP disguises for Puppy.) It's a tired but valid argument that it is what 90% of the global market find familiar, but an ever-diminishing proportion of those believe that it is "safe" or good value for their money. Can we have the flexibility to cater for the spectrum of user's 1) expectations and 2) level of interest in what lies "under the hood"? A self-organizing Puplet that artifus is suggesting is a great idea, I think, but perhaps it should be coupled with better online support, while the Puppy is still in the window and not yet taken for a test run by it's new owner. This could be facilitated by simple you-tube demonstrations about what each (official?) derivative does - in terms of basic hardware environment needed, what it's window environment looks like, how you can access and add preferred applications - and perhaps a highlight on what the (hush) d*v*l*p*r has done or added to achieve a particular goal? We can't let sneekylinux do all the you-tube spruiking, surely - even though he does a fantastic job. Perhaps we should provide an in-house video team, like the Documentation team?
And I'll get in now before my inevitable edit - apologies if this last idea has been already pitched elsewhere on the forum - I'm hoping to add to the discussion rather than keep chasing my own tail...
EDIT: Bugger. Guess my obsession to keep rewriting history happened again - Lobster started this video initiative over a year ago http://puppylinux.mirocommunity.org/category/. Mucho, mucho embarazo
Last edited by Puppyt on Sat 11 Jun 2011, 23:58, edited 1 time in total.
Search engines for Puppy
[url]http://puppylinux.us/psearch.html[/url]; [url=https://cse.google.com/cse?cx=015995643981050743583%3Aabvzbibgzxo&q=#gsc.tab=0]Google Custom Search[/url]; [url]http://wellminded.net63.net/[/url] others TBA...
[url]http://puppylinux.us/psearch.html[/url]; [url=https://cse.google.com/cse?cx=015995643981050743583%3Aabvzbibgzxo&q=#gsc.tab=0]Google Custom Search[/url]; [url]http://wellminded.net63.net/[/url] others TBA...
Someone mentioned "ALL" computers
Having done this for a very long time (talking about myself), why do we keep using the term "All" computers. And, at the same time, concern ourselves with 100MB CD limitation, while expecting that we can create a system which does everything on all computers. Wow. When I saw Barry, Kirk, and Playdayz add some specifics on PCs for each of these Distros (namely WARY, FATDOG, Puppy), that, to me, began a beginning round of identifying tested platforms which they support. We are too small to support ALL. And we should continue the vocabulary that started for PCs supported so that we can realistically achieve a reasonable objective at a very high level of quality. This doesn't mean restricting, it means classification!
When we post the word "ALL" we seem to be advocating something different from the current Puppy. It put down a statement (intent is there) We must understand that ALL will misguide and get misused by all the wrong methods. We shouldn't be afraid of saying "This is supported on these platforms. And while it may work on other plafforms its designed for these." This then take a level of professionaism to the community. And, allow us to focus. Yes, I have been able to deploy MS OSs on platforms they were never intended for, but, it was clear, to me, before I started what they intended. And, I could publish my finding. But even a behemoth like MS would never use "ALL". We shouldn't either.
Maybe, here, we SHOULD put a stake in the ground of what PCs (homes) the new Puppy is expecting to be boarded.
We are a smart group. Let keep the smartness growing.
Edited: Here's an example (but PLEASE don't use my words)
"I will support all PCs built after 2006 that are equivalent to Intel's P4 or greater, including Intel and AMD 64bit processors up to today. And, I will use all RAM that is reported to me from the system's BIOS. I also will use all peripheral reported after system startup within the Linux OS."
Yes, its still vague; but it does categorize the class machines I feel certain will have all adequate resources to easily run. Its provides enough specifics such that a clearer understanding of what this is designed for is extremely helpful for those planning to deploy.
Hope this helps.
When we post the word "ALL" we seem to be advocating something different from the current Puppy. It put down a statement (intent is there) We must understand that ALL will misguide and get misused by all the wrong methods. We shouldn't be afraid of saying "This is supported on these platforms. And while it may work on other plafforms its designed for these." This then take a level of professionaism to the community. And, allow us to focus. Yes, I have been able to deploy MS OSs on platforms they were never intended for, but, it was clear, to me, before I started what they intended. And, I could publish my finding. But even a behemoth like MS would never use "ALL". We shouldn't either.
Maybe, here, we SHOULD put a stake in the ground of what PCs (homes) the new Puppy is expecting to be boarded.
We are a smart group. Let keep the smartness growing.
Edited: Here's an example (but PLEASE don't use my words)
"I will support all PCs built after 2006 that are equivalent to Intel's P4 or greater, including Intel and AMD 64bit processors up to today. And, I will use all RAM that is reported to me from the system's BIOS. I also will use all peripheral reported after system startup within the Linux OS."
Yes, its still vague; but it does categorize the class machines I feel certain will have all adequate resources to easily run. Its provides enough specifics such that a clearer understanding of what this is designed for is extremely helpful for those planning to deploy.
Hope this helps.
Last edited by gcmartin on Tue 07 Jun 2011, 00:07, edited 2 times in total.
I would think that a "new" point for Puppy would be appropriate.
Old forms of computers to support would be a good discussion...but no longer a focus.
How old would you support?
And why support old kit?
It's a new world and happening fast....does Puppy need to support 486...586?...or move to the front and support the new processors?
Lot's out there is happening....get with the development and progress I figure.
Let's let the old go and embrace the new....Eric
PS..I didn't read the long posts..sorry All.
Old forms of computers to support would be a good discussion...but no longer a focus.
How old would you support?
And why support old kit?
It's a new world and happening fast....does Puppy need to support 486...586?...or move to the front and support the new processors?
Lot's out there is happening....get with the development and progress I figure.
Let's let the old go and embrace the new....Eric
PS..I didn't read the long posts..sorry All.
[color=darkred][i]Be not afraid to grow slowly, only be afraid of standing still.[/i]
Chinese Proverb[/color]
Chinese Proverb[/color]
Hi all
what brought me to puppy and linux was lucid 5, not because of any fancy gizmos or the distro it was compatable with, but it was a lean, light and simple fully functioning os. I just booted from the cd and away I went.
Now I am using a multisession dvd I can use my desktop on any of my pc's (even my old 800mhz amd with 512mb ram) would it be possible to make puppy write to a usb stick in the same manner (so it didnt automaticaly save a session and I could add a mobile swap file as well for pc's that do not have a swap file and limited ram) or perhaps a full install to a usb stick (making it truly mobile)
just some ideas
stripe
what brought me to puppy and linux was lucid 5, not because of any fancy gizmos or the distro it was compatable with, but it was a lean, light and simple fully functioning os. I just booted from the cd and away I went.
Now I am using a multisession dvd I can use my desktop on any of my pc's (even my old 800mhz amd with 512mb ram) would it be possible to make puppy write to a usb stick in the same manner (so it didnt automaticaly save a session and I could add a mobile swap file as well for pc's that do not have a swap file and limited ram) or perhaps a full install to a usb stick (making it truly mobile)
just some ideas
stripe
Hello,
Just use gparted to make partitions on the USB stick..
Make one Linux Swap, another fat32 for windoze access/sharing
And the third for Puppy...Ext2/3....
Install Puppy to it...
Just use gparted to make partitions on the USB stick..
Make one Linux Swap, another fat32 for windoze access/sharing
And the third for Puppy...Ext2/3....
Install Puppy to it...
Close the Windows, and open your eyes, to a whole new world
I am Lead Dog of the
Puppy Linux Users Group on Facebook
Join us!
Puppy since 2.15CE...
I am Lead Dog of the
Puppy Linux Users Group on Facebook
Join us!
Puppy since 2.15CE...
Hi Jay,
I would use an old USB stick for swap and put the "install" on another USB drive.
I used this successfully for a long time (2 sticks)..hell, still use an old Kingston 512 usb stick for swap (3 or 4 years?) dunno but a long time.
This started as a customer asked for a slim kit...it still works but she went to Apple...but it does still work...after many years and daily use...Eric
I would use an old USB stick for swap and put the "install" on another USB drive.
I used this successfully for a long time (2 sticks)..hell, still use an old Kingston 512 usb stick for swap (3 or 4 years?) dunno but a long time.
This started as a customer asked for a slim kit...it still works but she went to Apple...but it does still work...after many years and daily use...Eric
[color=darkred][i]Be not afraid to grow slowly, only be afraid of standing still.[/i]
Chinese Proverb[/color]
Chinese Proverb[/color]
Caneri Said:
Rather than look at a "new" point, maybe it's good to look the other way for the "old" point.
I have read a few recent posts saying things like "i've installed lupu 5.25 on my old p4 laptop with 256mb ram 60gig hdd" etc. In Puppy's "old" context, this is kinda new, don't you think?
For a Vista or 7 user, a PC that is older than 2 or 3 years is pretty old. Its almost as if its too old once its out of warranty! In that sense, all *our* ideas of old might be much further than others.
New for Puppy is within 5 years, but that is old for the Wintel types!!
Old computers have been a focus for Barry, and it would still be good to support them. i486 is VERY old, and probably too old to focus support on. I believe 01micko at one point said he had a 486 running puppy, or linux, in his shed, but reality is, machines of that age are useless for just about anything.I would think that a "new" point for Puppy would be appropriate.
Old forms of computers to support would be a good discussion...but no longer a focus.
Rather than look at a "new" point, maybe it's good to look the other way for the "old" point.
I have read a few recent posts saying things like "i've installed lupu 5.25 on my old p4 laptop with 256mb ram 60gig hdd" etc. In Puppy's "old" context, this is kinda new, don't you think?
For a Vista or 7 user, a PC that is older than 2 or 3 years is pretty old. Its almost as if its too old once its out of warranty! In that sense, all *our* ideas of old might be much further than others.
New for Puppy is within 5 years, but that is old for the Wintel types!!
-
- Posts: 229
- Joined: Wed 25 Aug 2010, 15:38
- Location: ISM Dhanbad, Jharkhand, India
Hi everyone,
On going through the thread I see that all want to add features to puppy but all are afraid of increasing size too much, well over 100MB.
I have a suggestion. Puppy should be built from source, but in a different way.
We can remove debugging symbols by filtering out '-g' option to gcc while compiling source packages.
This can be done in two ways.
1. Instead of running:
We can run:
2. We can write a wrapper script for gcc which will filter out '-g' option.
This will reduce size of puppy to almost half, if not, to two-third
Then we will get plenty of room for adding more stuff, and still stay around 100MB.
On going through the thread I see that all want to add features to puppy but all are afraid of increasing size too much, well over 100MB.
I have a suggestion. Puppy should be built from source, but in a different way.
We can remove debugging symbols by filtering out '-g' option to gcc while compiling source packages.
This can be done in two ways.
1. Instead of running:
Code: Select all
# ./configure [options] && make && make install
Code: Select all
# ./configure [options] && make install-strip
This will reduce size of puppy to almost half, if not, to two-third
Then we will get plenty of room for adding more stuff, and still stay around 100MB.
Have you tested this to see how much difference it actually makes? Cos I know a lot of compilers here already run the 'strip' command on their builds .. (So I do as well... )akash_rawal wrote:1. Instead of running:We can run:Code: Select all
# ./configure [options] && make && make install
2. We can write a wrapper script for gcc which will filter out '-g' optionCode: Select all
# ./configure [options] && make install-strip
I run 'strip --strip-unneeded *' on all the bins and libs
Is this the same thing?
[b][url=https://bit.ly/2KjtxoD]Pkg[/url], [url=https://bit.ly/2U6dzxV]mdsh[/url], [url=https://bit.ly/2G49OE8]Woofy[/url], [url=http://goo.gl/bzBU1]Akita[/url], [url=http://goo.gl/SO5ug]VLC-GTK[/url], [url=https://tiny.cc/c2hnfz]Search[/url][/b]
-
- Posts: 229
- Joined: Wed 25 Aug 2010, 15:38
- Location: ISM Dhanbad, Jharkhand, India
This reduces the size of binaries to almost half. Functionality is not affected at all.sc0ttman wrote: Have you tested this to see how much difference it actually makes?
I don't think, because I tried both methods on my boot loader manager and found that "./configure && make created a 201 KB executive. On compiling manually without '-g' option I got 90 KB executive. Using "strip --strip-unneeded" gives me even smaller 74 KB file. Dynamic loading of callback funcs still works in each case.sc0ttman wrote: I run 'strip --strip-unneeded *' on all the bins and libs
Is this the same thing?
Concerning higher performance Puppy on newer hardware--it seems that is an interest of several people--me included.
The Slackware binaries are built for i686. The Ubuntu binaries used in Lucid had been built for i386. To me, Slacko seems noticeably quicker. So that is happening.
01micko just mentioned in the Slacko thread possibly building a 2.6.39.1 kernel with PAE--I would like to see the latest kernel possible because I know from Lucid that after one year the kernel will seem long in the tooth.
There is Fatdog 64.
I have been getting 8-10% speed improvement from Firefox and Seamonkey by compiling them myself for my particular cpu. The newer your cpu the more improvement you will get. Some instructions are at http://www.murga-linux.com/puppy/viewtopic.php?t=67756
.
Seamonkey 2.1 is ***much*** faster than 2.0.14 even without any compiling tricks.
I am also interested in having current versions of the best Linux software available. The Lucid community has been a huge help in this. I look through Additional Software and find things periodically and make Lucid pets. The Lucid PPM contains packages that have been tested and configured for Lucid--I believe this will be important to the success of future Puppies. It's hard for one person to coordinate development *and* fill the PPM. Lucid is OK now, but I could have used a group who would find and make the latest software, make it into Lucid pets, and then test and send me the package. Perhaps at some point 01micko could use this.
The Slackware binaries are built for i686. The Ubuntu binaries used in Lucid had been built for i386. To me, Slacko seems noticeably quicker. So that is happening.
01micko just mentioned in the Slacko thread possibly building a 2.6.39.1 kernel with PAE--I would like to see the latest kernel possible because I know from Lucid that after one year the kernel will seem long in the tooth.
There is Fatdog 64.
I have been getting 8-10% speed improvement from Firefox and Seamonkey by compiling them myself for my particular cpu. The newer your cpu the more improvement you will get. Some instructions are at http://www.murga-linux.com/puppy/viewtopic.php?t=67756
.
Seamonkey 2.1 is ***much*** faster than 2.0.14 even without any compiling tricks.
I am also interested in having current versions of the best Linux software available. The Lucid community has been a huge help in this. I look through Additional Software and find things periodically and make Lucid pets. The Lucid PPM contains packages that have been tested and configured for Lucid--I believe this will be important to the success of future Puppies. It's hard for one person to coordinate development *and* fill the PPM. Lucid is OK now, but I could have used a group who would find and make the latest software, make it into Lucid pets, and then test and send me the package. Perhaps at some point 01micko could use this.