ksattic
8/12/03 10:33PM |
What is the reason for !Printers being very slow? As far back as I can remember, I always wondered why printing on RISC OS was slower than with Windows. Is it because the process requires a lot of FP operations?
Aside from that question, this is good news! I look forward to seeing a finished product! |
mavhc  (-1.0)
8/12/03 10:55PM |
Originally because RISC OS sent everything as graphics whereas Windows often used built in printer fonts. Now it's because x86 chips are billions of times faster. |
rod
8/12/03 10:58PM |
Excellent to hear this project is progressing - if it integrates well with the exising RISC OS !Printers system (as it sounds like it does), then this will hopefully be a major boost for the platform. |
rmacf
9/12/03 12:25AM |
I have my money ready and can't wait to try it. |
ksattic
9/12/03 1:01AM |
It will be free software, but donations would probably be very helpful. See [Link: www.mw-software.com] for more information. |
maikl
9/12/03 1:15AM |
I think it is a little pity that there will be the inbuild configuration for the resolution and an additional tool like GPSetup. So I would suggest to add a 'advanced options' button or something to the !Printers configuration so that it at least seems to be one program only. Now that the sources of !Printers are free - this should be easy to do for someone who knows more about programming than I do. A full integrated solution would of course be a further step forward.
I should add that I am really loking forward to GemPrint and this is only a small idea which should be easy to implement. Sorry for sounding a bit silly when trying to criticise a program that isn't even available. |
christ
9/12/03 1:50AM |
In reply to maikl:
The Printer configuration window does have an "Advanced..." icon, which can activate an external program based on the value of a system variable. There's a nice one for LaserJets, for example, which lets you set economy mode printing and the default text font.
I don't know whether Martin has used this interface, though. |
harmsy
9/12/03 8:30AM |
This is fantastic news - it sounds as if Martin already has it up and working. I don't think anyone would mind having to wait 24 minutes to get better printouts - think of what you could achieve in that time: boil a kettle, make the tea, milk the cow, pour the tea, drink the tea, wash up - then collect your print out. No really, this is top stuff (again) and Martin richly deserves support. WELL DONE. |
mrtd
9/12/03 9:21AM |
Yes, good stuff. I look forward to it, but will need a new printer to get the best out of it.
The one worry is that there are a lot of technical terms mentioned above that mean little or nothing to me. I have no idea whar "Drop Modulation technology" is, and the same went for "soft weaving" before I read the above. Is there going to be some sort of manual to explain the various "advanced" confiuration options that will be available.
Martin |
illudium (+0.1)
9/12/03 10:06AM |
This is starting to look really nice now. It's definitely time to buy a new photo printer, especially since I've just got a new camera.
In reply to christ:
Yes, one of the features of Printers+ is that it allows external configuration applications to be launched for the printers. I haven't seen anything make use of this yet though. It certainly makes sense for Martin to be supporting at least all the current versions of the Printers front-end though (1.64a, 1.67 and 1.91) as he seems to be doing. |
wuerthne (+0.1)
9/12/03 10:27AM |
To maikl and christ:
In reply to maikl:
I would very much like to get the Advanced button working but so far, I did not see any way to do this that works on all supported versions of !Printers. christ: I have never heard about any system variable. If there is one that works from 1.53 to 1.67 and 1.9x then that would be excellent news. Any more information on that? Of course, if there is an easy way to add support for external configuration programs under Printers+ I am more than happy to work together to make it easier to use. However, the first priority at the moment is to release something that works. |
wuerthne (+0.1)
9/12/03 10:34AM |
To mrtd:
Drop modulation technology means that the printer can print with drops of variable size. This is a major improvement in quality and one more feature our current native RISC OS drivers cannot make use of. "Soft weaving" basically means high-quality printing without stripes - again, our native drivers cannot do that (except possibly the now rather outdated SpaceTech drivers). Fortunately, you do not need to understand all the concepts involved. Just print and let Gimp-Print do everything for you. For experts, there is a lot of stuff to set up and control, e.g., colour balance, but you do not need that to get good printouts. |
wuerthne (+0.1)
9/12/03 10:39AM |
In reply to ksattic:
It is quite possible that it will not be free software. So far, I have not finally made up my mind. Parts of the project will certainly be free software, most notably GPDriver, which is based on Gimp-Print and hence has to be GPL'd. But it will be of little use without GemPrint. The GemPrint framework is more likely to be released in a more restricted way.
In reply to Re:
"!Printers is slow". If you look at the example figures above, you notice that the part of the printing process where !Printers is involved (the non-multi-tasking part) is only 35 seconds whereas the part handled by Gimp-Print takes 24 minutes (!). So, it is the exact opposite: !Printers is fast, but Gimp-Print is slow.. This, amongst others, is due to the fact that FP operations being used by Gimp-Print and that even an XScale is not as powerful as current X86 processors.
Martin |
wuerthne (+0.1)
9/12/03 10:42AM |
Incidently, discussions like the above should really happen on the RISC OS printing SmartGroup mailing list. How many people will look at this forum in a few days?
http://www.smartgroups.com/groups/riscos-printing
OR (purely e-mail based):
[Email: riscos-printing-subscribe [at] smartgroups.com ]
Martin |
em2ac (-1.0)
9/12/03 10:51AM |
I dont suppose anyone made an FP accellerator 2nd processor card for the RPC? |
sa110
9/12/03 11:14AM |
em2ac if a remember correctly there is a piece of software around that enable you to use the FPA from the x86 2nd proc cards. Don't know how good it is tho. |
rob
9/12/03 11:34AM |
In reply to sa110:
Nice idea, I think it was wam silence software. IIRC the overheads of calling the PC card made it slower than the FP emulator on a StrongARM. |
markee174 (-0.9)
9/12/03 11:35AM |
When you say it takes 24 minutes to print a page, what point is control handed back to the user. Does gimp-print multitask in the background? |
mrchocky (-1.0)
9/12/03 11:35AM |
sigh. I think em2ac really needs to be less excitable and do a little research before posting random comments. No, the software in question is no use on StrongARM systems, and there is no such accelerator card. |
illudium (+0.1)
9/12/03 12:58PM |
In reply to wuerthne:
[Link: support.riscos.com] section 3 has details on calling external config apps from Printers (this is the one using the system variable), and from the date of that it looks like I might have been wrong about it being a Printers+ only thing (I'm not near my RISC OS machine to check at the moment). |
em2ac (-1.0)
9/12/03 1:18PM |
In reply to sa110:
Cheers dude, now I have a use for the card! (been replaced with Very fast PC)
In reply to mrchocky:
dear god forbid me writting, and I have not much time to do research as I am a full time student, and work the rest of the time. |
dgs
9/12/03 1:37PM |
In reply to em2ac:
Sort of like that Chris Williams bloke, then? 
dgs |
IvanDobski (+1.0)
9/12/03 1:42PM |
In reply to Markee174:
"When you say it takes 24 minutes to print a page, what point is control handed back to the user. Does gimp-print multitask in the background?"
read the article above:
"Martin told us that printing such a page took about 35 seconds before the desktop began multitasking, followed by another 24 minutes and 80MB of RAM to convert the page for an Epson Stylus C84 at the highest quality setting. "
|
tamias (-1.0)
9/12/03 1:59PM |
"...there could be ways to separate the RISC OS printing layer from the main Pimp-Print code..." The mind boggles! |
hzn
10/12/03 5:22AM |
I like the sentence "Having a lot of RAM has never been a priority for RISC OS users, seeing as one can comfortably use a machine with only 64MB fitted"... Wow, until now I thought for RISC OS to run nicely some 16MB are fine. I even published GAG-News for quite some time with just 4MB on an A5000. Now having 512MB even with a RAM Disc and the like I just can't help having some 400-480MB freem memory |
| |
