Steady Progress towards Alpha 1
This weekend, the Haiku project has seen some nice leaps forward. Two items are especially noteworthy: Ingo Weinhold and Axel Dörfler have finally nailed bug 2059. This bug specifically prevented serious use of Haiku for anything else than testing, since it meant that the kernel could crash at any time, especially when there was heavy disk activity. All that was supposed to be written to disk at the time of the crash was lost. Luckily, due to prior fixes to the file system journaling and log replay, it didn't mean that your entire file system would be corrupted, but at least anything that you were working on at the time would be lost. So the fix for this particular bug is getting us much closer to our goal of a usable self hosting situation in which you can actually use Haiku for development. This is our most important goal to reach before we wanted to release the first alpha of R1.
The second noteworthy achievement is build system support for a mixed GCC4/GCC2 Haiku environment. It has been known for quite some time, thanks to the explorations of Haiku developer Michael Lotz, that it was possible to set up a GCC4 build of Haiku to run GCC2 applications or vice versa, by installing the respective libraries into certain places so that the correct versions are used for linking. What was missing was support in the runtime loader (the system component used to launch applications and link them to the shared libraries that they use) to do this automatically and on a system wide level. Also missing was support in the build system to effortlessly produce such a hybrid Haiku build. Both of these items have now been implemented by Haiku developer Ingo Weinhold. Also related to this, Michael Lotz had researched the stability issues that GCC4 builds of Haiku were suffering from some time ago and tracked them down to a problem in the specific GCC4 version that Haiku is using. They can be avoided by simply turning off a certain compiler optimization feature. All this combined means Haiku can use GCC4 itself while maintaining our stated goal of binary compatibility to the large pool of GCC2 applications in an automated and transparent fashion.
A few serious issues remain before we can release the first alpha. Some concern missing or buggy functionality that affect the self hosting goal with regards to the development tool chain. A completely native port of Subversion is the last item on this list. As far as I know, some bugs in the TCP implementation are preventing it, but progress is being made on this front as you read this. Formal testing is being conducted to make sure the entire tool chain will work correctly and reliably. Axel Dörfler is currently working on the device manager, the system component which manages everything concerning hardware and drivers. There are some issues with regards to hardware interrupts that, when fixed, will hopefully clear up some driver problems that can be experienced on certain hardware.
I want to conclude with a big "Thank You" to everyone who is helping with tracking and reporting issues in our bug tracker and to everyone providing patches and of course to the Haiku developers themselves! Personally, I am very excited about the progress that is being made. Thanks to everyone who is contributing towards this goal!
Comments
Re: Steady Progress towards Alpha 1
Having just re-read all of NoHaikuForMe's posts I feel in some ways he [gender assumption on my part...] is being unfairly represented here. He has made a couple of valid points, and none that I would actually dispute:
1. "Designed from the ground up" is potentially misleading, as Haiku have reused much of the design (at least high-level design - API, threading model, etc) from BeOS.
2. 64 bit is desirable for a modern desktop OS and will be even more so in the future, as lots of RAM and applications that take advantage of it become more common.
(3. Haiku don't claim anything special about media and they shouldn't because of a lack of good software)
That is actually all that he has said, and I agree with all 3 points. The problem is just the negative way they have been presented. Criticism is of course allowed, but there are more productive ways of presenting it. Of course maybe he just "threw a stick into the tree without any specific purpose", not to be helpful but just to generate a bit of a response - if so, well done!
A couple of the most blatant flame-bait statements he came out with do warrant a specific response as they represent unjustified attacks on Haiku without any facts to back them up:
The "desktop" claim is basically just a way to excuse swathes of missing features, such as in this case lack of support for modern CPUs.
False. The desktop claim is there to show the focus of everything the Haiku project does is for the desktop user. Any time there is a choice or a trade-off to be made it will be made with desktop use in mind. Overall the system will have a consistent feel, with a single-project mentality focusing on ease-of-use and responsiveness.
"Missing features" are explained by the fact that Haiku is still pre-alpha and has very limited resources when compared to, say, Linux and BSD.
For almost seven years now Haiku has appealed for and received developers, and money, plus a heck of a lot of free or very cheap exposure, to produce this supposed simple yet powerful system but there is little to show for it.
OK, so "little" is subjective, but I think the small team have achieved remarkable things since the project started. I can't see how anyone could boot up a nightly image and say there is little to show for the team's work.
It's basically a usable system now, can run the majority of R5 software, works better on newer hardware than R5, has improved networking, and can support self-hosted development. Neatly it can now also use the most recent GCC4 compiler whilst still retaining binary compatibility with GCC2-compiled applications.
As the title of this entire thread states, Haiku is making steady progress towards the first alpha release.
Simon
ps: Sorry for continuing the off-topicness of the thread. Most people are rightly very pleased with the progress made to date and looking forward to the future of Haiku. I can't wait to see the progress over the coming months and years and hopefully find some time to help out myself.
Re: Steady Progress towards Alpha 1
The trouble with this popular fairy tale is that none of it's true. Linus Torvalds didn't own a server, he'd bought himself a 386 desktop PC and decided to write an operating system kernel as a hobby. He wasn't interested in creating something big and professional, just a desktop system for himself and hopefully like-minded Minix users.
Thanks for the history lesson. However, I believe that there are a lot of influences on Linux these days, and a lot of it is to make sure Linux runs well as a server.
The point however is that when we write here that Haiku is designed for the desktop, we mean a couple of things by that. For example that the interface is meant to be responsive. Or that our goal is to include features only important for the desktop and to exclude features not directly important for that. The blob also doesn't make claims as to who designed it. Of course, much of it was designed by Be Inc, especially the API which dictated much of it's direct implementation. Some API is new and was designed by us, like the interface layout management. But - many parts of the OS are sufficiently far away from the API that they had to be designed by us, take the app_server for example. And then you gave the example of the network layer, though that is also an implementation of an existing design (TCP/IP and friends). It shows to me that your own definition of "design" is rather blurry.
However, that does not necessarily mean that the statement that "Haiku is designed from the ground up for desktop computing" is false or even misleading.
I don't think there's an honest dispute here, the Haiku blurb just doesn't reflect the actual project.
The blurb says pretty clearly that it is still in development. That alone means it states a goal. Both of these facts together mean the blurb reflects the actual project.
In this case Haiku contributors have become angry or disheartened because they read some of the posts in this thread that didn't agree with their presumptions. They can't believe that anyone could actually hold opinions different from their own, and certainly not that anyone could be in possession of contrary facts, which would make their own position untenable. But an alternative presents itself, perhaps the person specifically intended for them to become angry or disheartened. Such a person would be very malicious, but it is easier to believe that other people are frequently malicious than to admit to the possibility of being wrong...
To tell people who disagree with yourself that the problem is they cannot believe someone else has a different opinion is dragging the discussion to a useless meta discussion. Such arguments can be switched around easily. They are meaningless, since they don't defend your point, it's just a useless reasoning on why you are unable to convince your opponent. It may state the truth, but like I said, your opponent may claim the same about yourself, so what's the point?
If there are unreasonable "demands" they are probably built into Haiku's stated goals.
Why can't you accept the fact that we are not done yet and even say so in the blob that triggers your anger? Why can't you accept the fact that we have to distribute our limited resources and may be forced to keep certain stuff for later? You might find it unacceptable if some are missing from a "Desktop OS" (that's your right), while other might be able to live without them just fine. Simply ignore Haiku then.
I was originally interested in Mike Hancock's question which was rather woolly but worth answering in context.
Somehow you must have extended the scope of your answer, why else did you upset people?
I've found that by far the easiest way to appear to be well informed is to actually be very well informed and to stay away from topics about which either nothing certain is known or no-one important agrees. Plus as Kibo would be quick to point out, it's good to know how to use 'grep' and to be able to type.
Actually, it is enough to be good with words and make some claims which people judge too much effort to check up on. I have come across way to much "I have read somewhere that...[insert some claim about Haiku that I know to be false here]". Guess where people pick those up from...
I don't expect anything from Haiku.
Sorry, but that is not how you come across.
It's clear that a lot of BeOS users did expect stuff from Haiku. Some of them have been pleasantly surprised by how much has been done, many others (take Scot Hacker as an example) are mostly surprised that it's still going.
Another example of how good you are with words. Note the use of "Some" versus "many" and then the "backup" of the claim with a prominent name to appear well informed. I have read what Scot Hacker wrote, and I find it shameful that you translate it they way you do. How about a quote from him to back this up? (I can't be bothered to look up his article on BeTips, but I don't remember it like you put it.)
People who contributed their hard work, or even their money to the project have a right to expect some transparency (e.g. prompt availability of Haiku's financial and other corporate records)
Oh so, now it is suddenly about transparency again. Did you run out of arguments?
If you believe that any argument made in support of a claim about Haiku the project or Haiku the operating system is automatically "belittling" your work then I don't think I have much chance of following your line of thought at all, but it would explain why you're offended by me pointing out that Haiku largely just re-uses the BeOS design, despite the fact that you've never complained about Haiku's FAQ which says much the same thing.
I didn't talk about any claim or any argument or any critique. I was reacting to your argument, and that did come across as belittling our work. Why don't you stay on the topic what you said, and don't try to make this something general about whether or not we can take critique?
Basically, you take our stated goal, act as if it was a claim of what Haiku is (right now), and compare that with your own definition of what would fall under this goal. This is then your expectation of what Haiku should be in order to live up to these "claims". And then basically you tell us that with the time it took the project until now, and with all those resources it had, it should already live up to these goals, at least more so than it does. Which is my definition of belittling the work that I and other did and do.
Let's assume for a moment, that you worked on a project similar in complexity and scale as Haiku, and completed all the items that you find missing from Haiku. Even then I would find your argumentation and approach as arrogant and discouraging.
I would really love to know what motivates you to write these comments and what effect you are actually trying to have. I know that it takes a long time to follow other people responses and write your own responses as well as you do. I believe we will probably not get any further on the other points, but the answer to this last question really interests me: What are you trying to achieve? What would you consider a successful outcome of this discussion from your perspective?
Re: Steady Progress towards Alpha 1
Stippi, I knew such a guy at my previous work, and I know many more of those, this species seems to need to be right very much and will do so with lots of words and talking around it but they generally won't accept common truth, common sense and they lack intuition.
Such people are a waste of time to argue with since they don't really listen or want to learn to see from an other's point of view, they just want to feel that they are right.
The best way for us is to just ignore it and call it a troll. Sure if he or she has valid points, he or she could A: appologize even if for nothing, showing respect, we are all friends here B: get to the point.
Show some respect NoHaikuForMe or you will never be taken seriously and be rightfully be regarded as a troll.
If that's not enough you can always come to BeGeistert.
Re: Steady Progress towards Alpha 1
@NoHaikuForMe
No more tid for tad discussions for me: from the way you are approaching this discussion, I feel it could go on forever in rather pointless directions. Personally, I can't see anything positive coming out of this, so I would rather spend my time on something productive, both for myself and for Haiku, and I suggest you do the same. ;)
PS: Opinions and points of view are very subjective. Some here have already acknowledged that you do have some valid points. That does not mean that we all have to agree with you; Haiku is a big group of very diverse people after all.
PS2: FWIW, Haiku Inc. financial report for 2007 is being prepared as we speak and will be published in the near future.
Re: Steady Progress towards Alpha 1
May I just express my utter delight to be a part of this community! Thanks for handling this discussion in this dignified way. Can you imagine the flame-fest this thing would turn out somewhere else?
In the wise words of our "lost leader" Michael Phipps:
"To the community: remember that Haiku is a volunteer project. Respect and be kind to the volunteers. Imagine them standing next to you while you write emails and forum posts. Would you really say that if the team was in the same room as you? I have seen some of the nicest people that I have ever met say some of the least kind things with the anonymity of a computer in front of them. Also leave room for differing opinion."
Re: Steady Progress towards Alpha 1
False. The desktop claim is there to show the focus of everything the Haiku project does is for the desktop user. Any time there is a choice or a trade-off to be made it will be made with desktop use in mind. Overall the system will have a consistent feel, with a single-project mentality focusing on ease-of-use and responsiveness.
This trade off idea is interesting, for what it's worth there's never been much evidence that there is such a trade off to be made at the code level. NT engineers used to talk about such trade offs, but them someone annoyingly disassembled the code in question and showed that it's just obfuscated and actually their Workstation and Server products were identical code wise. This embarassment is avoided in newer NT builds by ensuring that each build vintage is only sold as either a "desktop" operating system or a "server" operating system and never both.
In terms of who does what and when it makes sense to have priorities, but the trouble is that Haiku is a volunteer project, so what actually gets done is decided by the individual volunteer regardless of the project's stated goals. Nobody is entitled, (and nor should they be) to tell Axel that he can't play with the eeePC and should get back to fixing filesystem bugs instead.
Actually, it is enough to be good with words and make some claims which people judge too much effort to check up on. I have come across way to much "I have read somewhere that...[insert some claim about Haiku that I know to be false here]". Guess where people pick those up from...
Unreliable and inflated claims are commonplace, you will see people (even those who've never run BeOS) commonly saying that BeFS is a "64-bit journalled filesystem" and that as a result you can store 18,000 petabytes of data in a file. This claim is mostly cited from Scot Hacker's BeView column, although he's not the only one to mention it, and comes originally from Be's own somewhat misleading White paper and other documents about BeFS.
The reality is not so impressive (though it was still adequate for Be's immediate needs). The 64-bit claim comes back to the size of the file offset data type. By tradition in most systems this is the same as the word size, ie 32-bits on most computers. Back in the 1980s it had been expected that 64-bit word sizes would be commonplace before anyone needed to deal with such big files. But we've already seen in this thread that didn't happen. This puts the maximum file size at 2^31 if you want to be able to seek within files using the C standard library. Big enough for a CD image, but not for full-length movies or other large media. A compile time flag was introduced to use a 64-bit data type on 32-bit machines, but Be instead of providing this flag elected to always use a 64-bit data type.
Meanwhile BeFS was to be re-designed for performance reasons. Dominic Giampaolo took inspiration from XFS and created a new filesystem. The theoretical maximum volume size was about 1000 petabytes (because of the limited size and number of allocation groups), but Dominic knew that there were many problems which would prevent it from scaling that far practically so he didn't consider this an important limitation.
Somehow the practical and even theoretical limits of BeFS were overshadowed by this single header file line defining a data type, and the rumour started that BeFS can handle file sizes up to 2^64 bytes. It's very persistent. Nearly always if someone's going to bring up BeOS they'll talk about the "64-bit journalled filesystem".
I have read what Scot Hacker wrote, and I find it shameful that you translate it they way you do. How about a quote from him to back this up? (I can't be bothered to look up his article on BeTips, but I don't remember it like you put it.)
“Have to say I’m with mneptok on this one. I don’t really see much compelling in using BeOS/Haiku today”
“And you think Haiku stands a snowball’s chance of getting into those markets?”
Oh so, now it is suddenly about transparency again. Did you run out of arguments?
I was trying to think what people could possibly be "expecting" from you that would be such a threat. Transparency is one of the few things people have actually asked for and it seems like a legitimate expectation to me. I guess a few people have asked for completely unrealistic things like running The Orange Box or whatever, but I don't think you need me to list those for you.
I would really love to know what motivates you to write these comments and what effect you are actually trying to have.
I would like people to be better informed so that they can make better decisions. Proponents of something tend to have, intentionally or not, a blind spot when it comes to that something's weaknesses. Someone earlier linked a article that shows some (poorly interpreted) results with 32-bit Photoshop on a system with 8GB of RAM. For several years Adobe employees claimed that any performance improvement from a future 64-bit Photoshop would be small and limited to those working with very large images - and so there was no reason to be annoyed that the current version wasn't available in 64-bit. Then, when they were able to announce that 64-bit Photoshop was at last under development suddenly they "discovered" that 64-bit Photoshop was noticeably faster than 32-bit builds of the same code for many operations, and an order of magnitude faster for large data sets with the same amount of RAM. No apologies were offered to all the people they'd shouted down previously.
Or take HD DVD, for some months prior to announcing that they had abandoned the format, some major studios were actually making preparations to switch to Blu ray because the sales figures for HD DVD were poor. Yet in statements to the press and public they continued to insist that HD DVD sales were meeting expectations and that they weren't switching. People who'd bought HD DVD players argued even more vociferously that HD DVD wasn't dead, or once that position was obviously hopeless, that Blu ray would flop too, with an HD DVD successor replacing both in a few years.
Re: Steady Progress towards Alpha 1
Abandoned by whom, pray tell? Please get off your high horse. It's basic dignity and goodwill that if you are not contributing to something you have no business putting in your two cents.
By our society, I am surprised you haven't encountered reviewers, critics, opinion columns and so on. Here's a popular web comic satirising your notion four years ago. Putting something into the public sphere opens it up to criticism by the hoi polloi by definition in our society. Whether you draw pictures, or play golf, whether you direct movies, or design clothes, people will offer their unsolicited and often unflattering opinions of your work to each other.
Re: Steady Progress towards Alpha 1
At first, I was responding to your other remarks, but I think it is pointless. This one is more like the essence anyways:
I would really love to know what motivates you to write these comments and what effect you are actually trying to have.
I would like people to be better informed so that they can make better decisions. Proponents of something tend to have, intentionally or not, a blind spot when it comes to that something's weaknesses.
I am sure that virtually everything out there has weaknesses. That's a pretty lame excuse to give up. Progress depends on being unreasonable, or at least on "hope". I am sure that there are people, who project hopes into what Haiku will be, to which it will not live up to any time soon or possibly never. It honors you that you are trying for them to be better informed. It doesn't follow that we Haiku developers have blind spots for Haiku's weaknesses or even our own capabilities and resources. Who do you think we are? Dumb people who don't know better and pretend to other people to be able to pull something off which we are incapable of doing and you need to protect the innocent? Please have some faith in other peoples intelligence and at least assume that from their point of view, it makes sense what they are doing and that they have at least some reason to hope for success. Obviously, we believe that Haiku will be the better basis to move forward with what we believe is a superior way of computing. We have to be stubborn and unreasonable at a certain level, if not for people like us, much less would be done in the world. If you don't find our project interesting then please just ignore us. If you think you need to warn other people of their likely disappointment, I am sure you are good enough with words to do so without discouraging us developers.
Re: Steady Progress towards Alpha 1
“Have to say I’m with mneptok on this one. I don’t really see much compelling in using BeOS/Haiku today”
“And you think Haiku stands a snowball’s chance of getting into those markets?”
Someone earlier linked a article that shows some (poorly interpreted) results with 32-bit Photoshop on a system with 8GB of RAM. .....
Ok, I did the Photoshop link but you missed the point completely. The review was on a very reputable and popular website that is known for benchmarking - Tom's hardware. 1) The point they were addressing was: Running 32-bit Software on 64-bit OS is still limited to within the 4GB limit. So, if you do install a 64-bit OS you need 64-bit software to take advantage of it ( get access to the full 8GB ). You said that the 64-bit Photoshop was much faster than the 32-bit but not how much memory there was in the systems. If the RAM was over 4GB then there is one problem, a 32 or 64-bit OS running a 32-bit program ( Photoshop ) can only work with 3 to 4GB. Whereas a 64-bit OS with 64-bit program can go beyond the 4GB limit. 2) Also, 64-bit speeds up complex computations because of more memory registers. So, assuming both systems were configured with 4GB. The 64-bit OS + 64-bit Photoshop would be quicker because there is lots of *complex* number crunching involved. And 3) if you run 32-bit software on 64-bit OS it'll be slower because you're using a compatibility layer as opposed to 32-bit on 32-bit OS ( or 64-bit on 64-bit OS ) which do not require compatibility layers.
You took this Photoshop example and never gave any specific details so that someone could dispute what you said. You didn't mention the amount of RAM on the systems which could have a big impact in itself. You didn't mention if both were using 64-bit OS which would have slowed down the 32-bit Photoshop. You also didn't take into account that Photoshop is one of the programs that *does* benefit from 64-bit on large data sets - even though Adobe claimed otherwise beforehand. Nor did you provide a link so that I or others could review the findings to discuss or challenge you any further.
The message was, if you run 32-bit software on 64-bit OS then you still hit the 4GB barrier & slower because you run through compatibility mode ( compared to 32-bit on 32-bit OS ). Most software available is 32-bit, and so only the few 64-bit programs out there will benefit from 64-bit OS.
***Maybe I haven't made this point clear enough, 64-bit will benefit those that use really memory intensive, number crunching programs and games.*** These are the only types of applications that benefit from 64-bit, like Photoshop using large data sets or Windows 3D games, etc. Will 64-bit make my Email? or Word? or IE/Firefox? or media player ( for mp3, mpeg & xvid playing )? Or messenger? etc. any faster. The answer is NO. Haiku is intended for general users and these people won't require 64-bit! Without 64-bit software, all a 64-bit OS will let you do is create very massive caches and multi-task between much more programs.
As for the quotes from Scot Hacker. Do you have links to prove these? Otherwise how do I know you quoted him directly? And, it happens that people can change their tastes and preferences over time. For Scot to say those would mean he has found an OS which makes him more than happy. Some people will be happy with Windows, others, Linux, others BSD, others BeOS/Haiku, etc. Just because one person loves a certain OS doesn't mean that everyone else should too. People can choose for themselves which OS they like or dislike.
PS So, you point out 1 application ( Photoshop ) that benefits from 64-bit, under a certain scenario - large data sets, which few will encounter. How about the vast majority of programs that do not require 64-bit? I guess those don't count in your books. Few programs take advantage of 64-bit so nothing is lost here. Plus, Photoshop still worked with 32-bit version, just somewhat slower.
Re: Steady Progress towards Alpha 1
As for the quotes from Scot Hacker. Do you have links to prove these? Otherwise how do I know you quoted him directly? And, it happens that people can change their tastes and preferences over time. For Scot to say those would mean he has found an OS which makes him more than happy. Some people will be happy with Windows, others, Linux, others BSD, others BeOS/Haiku, etc. Just because one person loves a certain OS doesn't mean that everyone else should too. People can choose for themselves which OS they like or dislike.
As much as I hate to add any fuel to this discussion, the Scot Hacker quotes are indeed true, and only a few months old in fact.
http://birdhouse.org/blog/2008/02/29/the-beos-tip-server-is-back/#comments
Either way, it's just opinion.
Re: Steady Progress towards Alpha 1
Hmm, well as part of a statitics project I am entering cpu price ghz fsb etc.... into a spreadsheet and I notice that at newegg.com they only have 4 32bit processors for sale out of 145 64bit processors and they have a very broad selection. IMO that is pretty interesting... if R1 is indeed 32bit only (not that big a problem IMO) then R2 probably ought to be 64 bit because by that time 64 bit PCs will really be thrown away as far as that goes.
Athlon 64 X2 4600+ 2.4ghz 0 L1 1024 L2 2000 FSB 2 cores 64/32 bit 65 watts $66 or 13.75 per GHz per core all of which is pretty beefy for the price.
That is pretty cheap a 64 bit sempron at 2ghz only cost $36 and that is faster than mine LOL time for teh upgrade XD
So, it is not that there is not any hardware to run a 64bit R1 in fact is is cheap (not free yet though :-) ). But, if it takes 2-3 years for R1 to be released then they might just be throwing these away by then (When Windows 9 comes out) so perhaps a dual build of haiku should be considered BeOS has supported multiple archs in the past and PPC is sort of in progress for Haiku isn't it? so What would be the implications supporting both 32 and 64 bit similar to Linux or perhaps a 32 bit kernel could load extensions like a driver into the kernel to support 64 on 64bit machines
On a side note, I have run out of ram before in Linux... rather easy to do in fact. I have say 512mb ram and I am loading 800x600 pngs (about 70kb each) to animate in gimp after about 240 images are loaded or less than 4 seconds of high quality frame rate video (30fps) i start swap thrashing like a crazy mon LOL... so that is not even hi def meaning that 1) gimp uses a ridiculous abmount of ram and 2) if haiku is to be a multi media powerhouse at any point it will require a truck load of ram not for streaming multimedia but for actual editing and manipulation also at one point I rendered an image (50mb png) that was actually too large to load into ram why render that large you ask .... well you get better details that you can see otherwise... and yeah it was ridiculously huge :-) like 20000x15000 I think it was
There is one thing that could be done perhaps.... better support for dynamic access of large files MS actually bought a very interesting company that had a tech that might actually be nice basically they can view thousands of images simultaneously manipulating them zooming in thousands of times in seconds that kind of multi media power is really needed not that MS is even capable of even doing anything even half witted with it..... after a little googleing i have found the software MS calls it Photosynth originally it was called SeaDragon so go youtube the video of it and be boggled very impressive use of math (matrices among other things i believe) to do this : http://www.google.com/search?q=image+viewing+microsoft+technology
I believe that the API will be open for that software perhaps a similar thing to wine or gnash could get off the ground
Well I hope that post was interesting enough.... :-) albeit very broad in subject matter
Re: Steady Progress towards Alpha 1
Any software for media editing can be naive or smart about loading stuff into RAM. Unfortunately, many don't strike a good balance. The GIMP example you gave just means that GIMP was not written with that use in mind. If enough people complain, maybe they fix the software to do on-demand loading. I guess present day hard drives will be capable of loading 70KB PNGs fast enough for animation. And once that is done, you will be able to play back *any* length of animation regardless of 32 or 64 bits. It's just a matter of how smart the software is. I will give you two more examples: A couple of weeks ago, I used OpenOffice to edit a large document with lots of pics. The software became completely unusable on a quad core with 2 gigs of memory until I found the setting to change the image cache. Unfortunately, there was an upper limit to the cache (256 MB?) (the default was ridiculously low), so I could fix the problem only halfway.
Another example is Sony Vegas. I think it is a damn good video editor, especially considering the price. But it is incredibly stupid about caching the thumbnails on the timeline. You can imagine that creating these thumbnails from HD footage, especially while playing it back at the same time, is an expansive operation. Yet Vegas will only cache barely as much thumbnails as are visible on twice the width of the current timeline, so that it has to constantly regenerate the thumbs when you scroll left or right. It totally spoils the workflow and there is no setting to increase the thumbnail cache. How incredibly stupid considering the tiny amount of memory that a thumbnail needs compared to everything else the software does...
So I am just saying that multi media software needs to be smart. You can hit all sorts of limits with multimedia software, another example would be maximum open files. So you have to code it smartly and be wise about what to load into RAM and cache, and what to load only temporarily during playback. If you hit a limit at 4GB, you will hit it in twice the time already with 8GB in case the software is 64 bits. So you can see how 64bits would help you very little if the software is naive about RAM usage.
Re: Steady Progress towards Alpha 1
Hmm, well as part of a statitics project I am entering cpu price ghz fsb etc.... into a spreadsheet and I notice that at newegg.com they only have 4 32bit processors for sale out of 145 64bit processors and they have a very broad selection. IMO that is pretty interesting... if R1 is indeed 32bit only (not that big a problem IMO) then R2 probably ought to be 64 bit because by that time 64 bit PCs will really be thrown away as far as that goes.
I can't believe we're still talking about this. Stephan has said it simply enough - yes, we know 64 bit is important but right now it's not a top priority. That's because there's still lots to do to get Haiku finished on 32 bit hardware - and 64 bit will be able to run that too.
There is one thing that could be done perhaps.... better support for dynamic access of large files MS actually bought a very interesting company that had a tech that might actually be nice basically they can view thousands of images simultaneously manipulating them zooming in thousands of times in seconds that kind of multi media power is really needed not that MS is even capable of even doing anything even half witted with it..... after a little googleing i have found the software MS calls it Photosynth originally it was called SeaDragon so go youtube the video of it and be boggled very impressive use of math (matrices among other things i believe) to do this : http://www.google.com/search?q=image+viewing+microsoft+technology
I'm doing a PhD in computer vision at the moment, and the stuff they do registering images and getting the camera position is actually not too difficult. Interesting they never show a single 3D model built from the photos, that's a harder problem (different lighting, occlusion, gaps, etc). I hadn't seen the SeaDragon stuff, that's pretty neat, thanks for the link.
Re: Steady Progress towards Alpha 1
@NoHaikuForMe
I can very much agree in the last part, IMO BeOS was overhyped, some tricks it uses are more "magical" then others but I think the idea and direction that have been set, are a good.
I hope Haiku will not suffer the same marketing buzz but you know how it goes... IMO it's quite natural, how to inform other people that you find the product great? Just tell them how you feel about it, will it be an accurate reflection of the real benefits and specs of the product? Probably not. Does it matter? Depends...
Still weren't you expecting more transparency and direction from 'something Haiku management' first and now you don't expect volunteers like Axel to do bug hunting because that should actually be a higher priority instead of playing with an eeePC?
Re: Steady Progress towards Alpha 1
Yeah well, Scot has IMO always been a dreamer, sometimes that can be useful, eg. inspiring others to create what he envisioned.
Re: Steady Progress towards Alpha 1
Scot is firmly on board the mac camp. I am much more interested in mac as competition than linux (which tries to copy windows) as Apple seems to understand what "designed for desktop" really means. That's why Apple has made real progress in marketshare while linux is still 2% or so despite all the hype.
But Scot has no problem with mac being closed source: he is consistent as he had no problem with BeOS being closed source.
I do have a problem with a closed source OS. I also don't like the idea of paying for expensive OS updates and the fact that you can only use mac os x on apple hardware. Apple is also competing with its own developers. It promises to be just as bad as Microsoft except more stylish.
Re: Steady Progress towards Alpha 1
yeah i know gimp is stupid at handleing animation :-) ... perhaps cinelerra does better..
What I was trying to do (aside from redirecting this discussion to legit topics ) was get across that software can be dumb not that I conveyed that very well. So would it be possible to implement intellligent media handling in the media kit like optimised cacheing or even SeaDragon like stuff? or is it beyond it's scope? I don't think that it would be too far fetched not to mention really usefull perhaps even changeing the way image browseing works in haiku. Hopefully if it were implemented to do that kind of thing it would curb some of the stupid things programs do sometimes like nom ur ram
Re: Steady Progress towards Alpha 1
i can't express how glad i am to read about the gcc2/gcc4 interoperability advancements. haiku is getting closer and closer to the day we, little fun-loving devs, will be able to mind our work in our favourite clover field. *smells the green*
as about people complaining about 'modern performance', little do they know that a modern compiler support is way more important for the overall performance of the application base than something like 64-bit address space. but oh well, they say the devil was in the details, and the level of holiness of some of the posters here is astounding ; )
Re: Steady Progress towards Alpha 1
Whether you draw pictures, or play golf, whether you direct movies, or design clothes, people will offer their unsolicited and often unflattering opinions of your work to each other.
And did I ever claim otherwise? Nope. I simply argued that criticism which is not helpful to the work being criticized exhibits a complete lack of tact. Critics are unpopular for a reason. :)
Re: Steady Progress towards Alpha 1
And did I ever claim otherwise? Nope. I simply argued that criticism which is not helpful to the work being criticized exhibits a complete lack of tact. Critics are unpopular for a reason. :)
Yes, you explicitly claimed that people who aren't contributors have "no reason (or right) to criticize". When I pointed out that this notion had been rejected by our society you became indignant, and even insisted that people have "no business" engaging in this perfectly ordinary activity.
I'm glad that you've come to your senses and won't be setting off to play cultural King Canute.
Re: Steady Progress towards Alpha 1
@NoHaikuForMe
I'm tired of going in unproductive circles with you. Just because there exists critics does not mean non-constructive criticism is socially acceptable on a personal level. Let's say I hold an art exhibit of my work, and you come in off the street and tell me my paintings suck; I would suspect most people would consider that offensive.
Re: Steady Progress towards Alpha 1
@Jonathan Freeman
well who cares what NoHaikuForMe says? obviously you have taken it personally Jonathan you even went to the trouble to put italics in your post... anybody with half a brain (possibly less) can tell that Haiku is being developed in an intelligent manner just look at the site it looks professional. Smart people don't decide about something based on one source especially when it comes accross in such an impolite manner
and by the way i would make sure my art didn't suck before i put it on exhibit :-) no worries though Haiku doesn't suck especially for pre-alpha.
admitting your wrong bites big (well not really but it seems that way) time but I have to admit i have said some stupid things before and even tried to back them up matter of fact i did just a little of that over in the opengl thread not that i was all wrong but still... personally i find this whole argument silly and immature and I'm not all that mature actting myself sometimes but this is rediculous (me repeating myself not this thread XD )
But anyway HURRAY FOR GCC4 HYBRID SUPPORT IZ AWESOME! since i think forgot to say that earlier you guys are awesome and indeed inspire me a programmer perhaps I can do something for haiku latter on :-)