- Pharo-dev - lists.pharo.org

Re: [Pharo-project] do you know what is this?

by Esteban Lorenzano

Hi, On Apr 17, 2013, at 11:23 AM, kilon <thekilon(a)yahoo.co.uk> wrote: > Anyway, my only objection is the fact that we will have twice the code to > maintain and more to worry about. let me explain this little thing a bit... Do you think that code doesn't need to be maintained and taken care how it is right now? I will draw you a picture of how it works this now /----------------------------------------------\ | "Pretended wonderland" (image) | =========================== | Lots of things: I/O, FFI, etc. | ----------------------------------------------- | Virtual Machine (Cog, Stack, Int) | \----------------------------------------------/ This is your world right now. You work on "pretended wonderland" side, assuming that everything below will work and hence you "don't need" to maintain what is below wonderland, isn't? Well, wrong. Everything need to be maintained. What you do not see too (you do not see it, but it is there, oh yes). The difference is: - in wonderland, everybody can try to fix things (even id they fail, it is there to play/change/adapt) - in levels below, there are just a few people on earth who actually can maintain that: Igor, myself and some others, but is for sure less than ten people. In fact, what happens now is that I have to maintain not two, but three different versions of the same thing (for each platform), in two different languages (C and Objective-C) and three different sets of libraries. And yes, I "personalize" and use the "I" instead "we" because most of the times is me, with some really valuable from Igor and Camillo (but well, they do Athens and a PhD respectively, so many times is just a one-man-work). So, what actually happens now is that problems stack waiting for a solution, a solution which btw will never be of total satisfaction of users. As an example, let's see the display plugin (which of course is not a plugin at all, if you remove that from vm, it stops to work). Display plugin is based in BitBlt, who creates/transforms bitmaps and renders them into screen. Forget about 2D vectorial graphics (or any thing that can bebefict of using of GPU). This technology is obsolete and retina displays on mac prove that: there is no way to do it work fine (that means: good rendering and good performance). So we need to change that. How to do it? There are two basic approaches possible: The traditional one says that some "gurus" should write a new plugin (ideally backward compatible with the previous one). Then you have an updated plugin (and probably a perfect one, if you don't want to maintain it), until next time technology changes and then the gurus can be summoned again to do their magic. As you can see, since our community of potential gurus is (with luck) ten persons in the world, that does not scales at all. Then it is our approach: move to the image all the things that can be moved. Then you will have this: /----------------------------------------------\ | "Pretended wonderland" | ----------------------------------------------- | Abstract system layer in image | | (that most of people will never | | see anyway) | =========================== | Virtual Machine (Cog, Stack, Int) | \----------------------------------------------/ With NativeBoost we can achieve same performance that we achieve using plain C. And we have an incredible huge extra bonus: it is there so everybody can try to modify/fix/update if needed, not just the gurus. And we also have much more advantages to show: this is moving also the low level programing to "pharo way", and as I showed in my silly videos, is bring the "immediate feedback" experience also to that level (something that you could never achieve with older way of thinking) And finally... been in image does not need that you will need to deal with that. How many of you know of the internals of Zinc/Zodiac or the Weak Registry? (just to put a couple of examples). Most of us just take it and use it. Our path is to bring to the whole community the power to change everything in your environment. To change it live, in the pharo way. And some times I feel that it is a power that the community is afraid to take :) lose your fear: having that power feels good :) And again: do you still think there is no need for maintain low level stuff? Think again. cheers, Esteban

April 17, 2013

Re: [Pharo-project] How to browse a selected method in code viewer

by Marcus Denker

On Apr 17, 2013, at 11:51 AM, kilon <thekilon(a)yahoo.co.uk> wrote: > Is it a bug ? If I select a class and do alt+b it works , if I select the > entire message it does not . For example Yes, there is a bug report there already since Apr 9, 2012: https://pharo.fogbugz.com/f/cases/5628/More-permissive-navigation-Shortcuts… Marcus

April 17, 2013

Re: [Pharo-project] do you know what is this?

by Esteban Lorenzano

Hi Phil, On Apr 17, 2013, at 9:26 AM, "phil(a)highoctane.be" <phil(a)highoctane.be> wrote: > Regarding this, of the key points of Pharo is that most of the system is able to be looked at from a lot of angles in source code form. I do not understand this... why do you think that now you can do that and with an all-in-image approach you cannot? > > Removing that ability is not going to be nice. why do you think that this will remove it? Esteban > > But this doesn't mean we can't go the other way around of course, allowing it to embrace new technical possibilities. But as extensions to a core that stays under full control. > > At the moment, I've been tinkering with the idea of integrating a 3D demo engine (DirectX based) with Pharo. > > The engine will get an API and being scripted from the Pharo side. > > The engine can currently do things like: http://www.youtube.com/watch?v=oRnogMokS_8 > > Phil > > Phil > > > 2013/4/17 Esteban Lorenzano <estebanlm(a)gmail.com> > Hi, > > This is just the cairo library rendering in a quartz format. No native quartz access yet, so... > > 1) yes... is 100% compatible with Cairo (because it is Cairo :) > 2) no GPU, no OpenGL, just optimized rendering (one of this days I will write the novel "how we render world canvas nowadays"... it is a thriller end-of-the-world story) > > we are thinking with Igor if a quartz native renderer has sense right now. Of course, in the long term, it has completely sense, but atm, I have the feeling that other more portable renderer (like an OpenGL renderer) is better investment. > Anyway, we still have some other stuff to do before: > > - finish athens (including new text model) > - finish bridge > - test same with linux, windows > etc. > > On Apr 17, 2013, at 7:50 AM, Denis Kudriashov <dionisiydk(a)gmail.com> wrote: > >> Hi >> >> 2013/4/17 Fernando Olivero <fernando.olivero(a)usi.ch> >> NICE work Esteban! >> >> Does the AthensQuartzSurface, support all Athens related code? (paths, >> paints,etc..) I will be more than happy to move away from the software >> rendering of the current cairo backend. >> >> Is Quartz work by GPU? is it use openGL? >> >> >> Fernando >> >> >> >> On Tue, Apr 16, 2013 at 8:41 PM, Esteban Lorenzano <estebanlm(a)gmail.com> wrote: >> > not to say about efficiency. >> > >> > I've been doing some cpu cycle tests. >> > >> > Rendering a 1600@1200 gradient to world canvas (repeat each 50ms): ~90% cpu consumption >> > >> > Same test, but sending it direct to window canvas using bridge: ~40% cpu consumption >> > >> > so... we are in the good path :) >> > >> > Esteban >> > >> > On Apr 16, 2013, at 8:28 PM, Igor Stasenko <siguctua(a)gmail.com> wrote: >> > >> >> On 16 April 2013 19:09, kilon <thekilon(a)yahoo.co.uk> wrote: >> >>> And you guys then say that pharo is not macos first citizen >> >>> >> >>> ha >> >>> >> >>> ha >> >>> >> >>> and >> >>> >> >>> ha >> >>> >> >>> I am a macos user , I love my imac and macos, but my vote goes to cross >> >>> platform. >> >>> >> >>> Still another great library that is more than welcomed, definitely cant do >> >>> any harm ;) >> >>> >> >> >> >> By saying that we don't need Cairo on MacOS i meant following: >> >> >> >> Athens designed to support multiple backends. >> >> It is out of question, that better to use most suitable backend, >> >> available on current platform. >> >> >> >> But apart of it stays ObjC bridge. Which would allow us to speak with >> >> ObjC-runtime >> >> (and Mac VM using it) directly. >> >> >> >>> >> >>> >> >>> -- >> >>> View this message in context: http://forum.world.st/Fwd-do-you-know-what-is-this-tp4681962p4681975.html >> >>> Sent from the Pharo Smalltalk mailing list archive at Nabble.com. >> >>> >> >> >> >> >> >> >> >> -- >> >> Best regards, >> >> Igor Stasenko. >> >> >> > >> > >> >> > >

April 17, 2013

Re: [Pharo-project] do you know what is this?

by Igor Stasenko

On 17 April 2013 08:34, Esteban Lorenzano <estebanlm(a)gmail.com> wrote: > Hi, > > This is just the cairo library rendering in a quartz format. No native > quartz access yet, so... > > 1) yes... is 100% compatible with Cairo (because it is Cairo :) > 2) no GPU, no OpenGL, just optimized rendering (one of this days I will > write the novel "how we render world canvas nowadays"... it is a thriller > end-of-the-world story) > > we are thinking with Igor if a quartz native renderer has sense right now. > Of course, in the long term, it has completely sense, but atm, I have the > feeling that other more portable renderer (like an OpenGL renderer) is > better investment. Right. In a longer term, i think OpenGL is a way to go, to get access to hardware-accelerated renderer. It is , however a lot more work comparing to providing thin layer on top of existing vector-graphics library, because OpenGL operates with much more primitive entities: triangles. > Anyway, we still have some other stuff to do before: > > - finish athens (including new text model) > - finish bridge > - test same with linux, windows > etc. > > On Apr 17, 2013, at 7:50 AM, Denis Kudriashov <dionisiydk(a)gmail.com> wrote: > > Hi > > 2013/4/17 Fernando Olivero <fernando.olivero(a)usi.ch> >> >> NICE work Esteban! >> >> Does the AthensQuartzSurface, support all Athens related code? (paths, >> paints,etc..) I will be more than happy to move away from the software >> rendering of the current cairo backend. > > > Is Quartz work by GPU? is it use openGL? > >> >> >> Fernando >> >> >> >> On Tue, Apr 16, 2013 at 8:41 PM, Esteban Lorenzano <estebanlm(a)gmail.com> >> wrote: >> > not to say about efficiency. >> > >> > I've been doing some cpu cycle tests. >> > >> > Rendering a 1600@1200 gradient to world canvas (repeat each 50ms): >> > ~90% cpu consumption >> > >> > Same test, but sending it direct to window canvas using bridge: ~40% cpu >> > consumption >> > >> > so... we are in the good path :) >> > >> > Esteban >> > >> > On Apr 16, 2013, at 8:28 PM, Igor Stasenko <siguctua(a)gmail.com> wrote: >> > >> >> On 16 April 2013 19:09, kilon <thekilon(a)yahoo.co.uk> wrote: >> >>> And you guys then say that pharo is not macos first citizen >> >>> >> >>> ha >> >>> >> >>> ha >> >>> >> >>> and >> >>> >> >>> ha >> >>> >> >>> I am a macos user , I love my imac and macos, but my vote goes to >> >>> cross >> >>> platform. >> >>> >> >>> Still another great library that is more than welcomed, definitely >> >>> cant do >> >>> any harm ;) >> >>> >> >> >> >> By saying that we don't need Cairo on MacOS i meant following: >> >> >> >> Athens designed to support multiple backends. >> >> It is out of question, that better to use most suitable backend, >> >> available on current platform. >> >> >> >> But apart of it stays ObjC bridge. Which would allow us to speak with >> >> ObjC-runtime >> >> (and Mac VM using it) directly. >> >> >> >>> >> >>> >> >>> -- >> >>> View this message in context: >> >>> http://forum.world.st/Fwd-do-you-know-what-is-this-tp4681962p4681975.html >> >>> Sent from the Pharo Smalltalk mailing list archive at Nabble.com. >> >>> >> >> >> >> >> >> >> >> -- >> >> Best regards, >> >> Igor Stasenko. >> >> >> > >> > >> > > -- Best regards, Igor Stasenko.

April 17, 2013

Re: [Pharo-project] do you know what is this?

by Igor Stasenko

On 17 April 2013 11:23, kilon <thekilon(a)yahoo.co.uk> wrote: > Estaban correct me if I am wrong but AFAIK MACOS GUI is based on opengl and > it is hardware accelerated. I hear the same applied for Linux GUIs too in > large. > Not at all. It is same story as with Athens: - GUI uses Cocoa framework. Sure you can use OpenGL or Quartz for rendering, but in most cases you just have a canvas where you draw to using common protocol. So, you don't need to think whether it is opengl-based or quartz or cairo. > Anyway, my only objection is the fact that we will have twice the code to > maintain and more to worry about. As I said I am on macos , I am a mac > developer (not pro) still ObjC never made sense for me since it locks me > down to a specific platform. It does however make sense to have platform > specific code for providing better and smoother support for that platform. > So its definetly good to have and thank you for your efforts. > Right. The objective for ObjC bridge is smoother integration with mac platform. It has little to do with Athens but more about communicating with underlying platform using language it understands. > Speedwise , I know the subject on optimisation is a huge one and there > thousands of ways to speed up specific code. Its a more a practical one, > since not to many people work on athens we need to choose one way that makes > life easier for everyone and especially people who develop Athens. > > Part of the sucess of Cairo is that is an easy to understand little library > that is very well documented. I would like to see Athens maintain that. > We will get there. :) > > > -- > View this message in context: http://forum.world.st/Fwd-do-you-know-what-is-this-tp4681962p4682078.html > Sent from the Pharo Smalltalk mailing list archive at Nabble.com. > -- Best regards, Igor Stasenko.

April 17, 2013

Re: [Pharo-project] How to browse a selected method in code viewer

by kilon

Is it a bug ? If I select a class and do alt+b it works , if I select the entire message it does not . For example In Athens-Cairo-Paints packages AthensCairoPatternPaint>>intialize in line NBExternalResourceManager addResource: self data: handle. if I select "NBExternalResourceManager" alt+b works and browses the relevant class If I select "NBExternalResourceManager addResource: self data: handle." and hit alt+b it does nothing Is that a bug and can you reproduce that bug ? if yes in both accounts then I am more than happy to create the bug report. -- View this message in context: http://forum.world.st/How-to-browse-a-selected-method-in-code-viewer-tp4681… Sent from the Pharo Smalltalk mailing list archive at Nabble.com.

April 17, 2013

Re: [Pharo-project] DateAndTime bug: we should do something about it

by Henrik Johansen

On Apr 16, 2013, at 7:50 PM, Eliot Miranda wrote: > > > On Tue, Apr 16, 2013 at 2:56 AM, Igor Stasenko <siguctua(a)gmail.com> wrote: > On 16 April 2013 11:32, Henrik Johansen <henrik.s.johansen(a)veloxit.no> wrote: > > > > On Apr 15, 2013, at 4:50 PM, Igor Stasenko wrote: > > > >> On 15 April 2013 16:43, Igor Stasenko <siguctua(a)gmail.com> wrote: > >>> On 15 April 2013 16:32, Henrik Johansen <henrik.s.johansen(a)veloxit.no> wrote: > >>>> > >>>> On Apr 15, 2013, at 3:25 PM, Igor Stasenko wrote: > >>>> > >>>>> On 15 April 2013 12:14, Henrik Johansen <henrik.s.johansen(a)veloxit.no> wrote: > >>>>>> > >>>>>> On Apr 13, 2013, at 12:16 AM, Igor Stasenko wrote: > >>>>>> > >>>>>>> > >>>>>>> But that's fine.. now look at > >>>>>>> > >>>>>>> secondsWhenClockTicks > >>>>>>> > >>>>>>> "waits for the moment when a new second begins" > >>>>>>> > >>>>>>> | lastSecond | > >>>>>>> > >>>>>>> lastSecond := self primSecondsClock. > >>>>>>> [ lastSecond = self primSecondsClock ] whileTrue: [ (Delay > >>>>>>> forMilliseconds: 1) wait ]. > >>>>>>> > >>>>>>> ^ lastSecond + 1 > >>>>>>> > >>>>>>> that is complete nonsense. Sorry. > >>>>>>> > >>>>>>> This code relying on primSecondsClock resolution, which is..... (drum > >>>>>>> roll..... ) > >>>>>>> 1 second.. > >>>>>>> > >>>>>>> then it is combined with millisecondClockValue , as you see later to get > >>>>>>> system time with millisecond precision.. > >>>>>>> > >>>>>>> I am not genius in math and physics.. but even i understand that if > >>>>>>> you measurement has error X > >>>>>>> you cannot get more precision than X, even if you combine it with > >>>>>>> another measurement with higher precision. > >>>>>>> > >>>>>>> (But i can be wrong with that.. if so, please explain why) > >>>>>> > >>>>>> Because, as Levente said in code, resolution != accuracy. > >>>>>> When you measure a single sample, sure, that value has an error of max resolution / 2. > >>>>>> But when you measure when that value _changes_, you have an error of the underlying accuracy / 2. > >>>>>> Clock s value: 1 2 > >>>>>> resolution: |_______|_______| > >>>>>> accuracy: ||||||||||||||||||||||||||||||||||| > >>>>>> ^ > >>>>>> Here's to hoping the above ascii art(?) worked! > >>>>>> > >>>>>> So no, that method is not complete nonsense. > >>>>>> > >>>>> Okay. > >>>>> Now i rewriting the code which uses new primitive (primUTCMicrosecondClock), > >>>>> and then we will no longer need this startup logic nor wrap around > >>>>> check & other gory details... > >>>>> just use a primitive. > >>>> > >>>> Sounds clearer, yes. > >>>> > >>>> However, be aware, at least on my VM: > >>>> [Time millisecondClockValue] bench '34,600,000 per second.' > >>>> [Time primUTCMicrosecondClock] bench '12,400,000 per second.' > >>>> > >>>> Which in my case translated to that in a Time2 class with #now defined: > >>>> now > >>>> > >>>> | microSecondsToday | > >>>> microSecondsToday := self primUTCMicrosecondClock \\ 86400000000. > >>>> ^ self seconds: microSecondsToday // 1000000 nanoSeconds: microSecondsToday * 1000 > >>>> > >>>> and changing the relatively low-hanging fruit in standard Time (removing Duration creation and other crap): > >>>> > >>>> seconds: seconds nanoSeconds: nanoCount > >>>> "Answer a Time from midnight." > >>>> > >>>> ^ self basicNew > >>>> seconds: seconds nanoSeconds: nanoCount > >>>> > >>>> To benchmarks that looked like this: > >>>> "Old code" > >>>> [Time now ] bench '7,330,000 per second.' '7,280,000 per second.' '7,170,000 per second.' > >>>> "Using microsecondClock every call" > >>>> [Time2 now ] bench '2,280,000 per second.''2,300,000 per second.' '2,180,000 per second.' > >>>> > >>>> Also note, Time2 lacked translation to Local time, which would be needed for consistency with Time now. > >>>> > >>> > >>> Well, i am not started modifying Time now, (will start in couple minutes ;) > >>> but here results of DateAndTime now > >>> before and after: > >>> > >>> before: > >>> > >>> > >>> [ DateAndTime now ] bench '316,000 per second.' > >>> > >>> after: > >>> > >>> [ DateAndTime now ] bench '414,000 per second.' > >>> > >>> > >>> with Time now, i think even if it will be "2,300,000" answers per second, > >>> it is fairly acceptable , when accuracy of primitive is 1 microsecond, > >>> you actually don't need more than 1 million answers per second, or you > >>> will receive same answers. > >>> > >>> But aside of it, may i ask, why performance is so important/critical > >>> for those? What is use case? > >>> > >> > >> everything is good in comparison: > >> > >> [ self yourself ] bench '58,800,000 per second.' > >> > >> does nothing and just 25 times faster.. than Time now, > >> so what is the point? :) > > > > The point was raising awareness that performance differences are also worthy of measuring when doing a reimplementation, not just code cleanliness. > > A 3x decrease in performance is worth at least consideration as part of a reimplementation, even when the conclusion is "it's plenty fast enough for most uses anyways". > > > > Not to mention, "Made X Y% faster" looks better in the release notes than "Restructured the internals of X" :) > > > RIght. > So, it is 3x slower , 1000x more accurate (milli vs micro), do not > requires startup code > do not need to account for counter wrapping and day change => much > cleaner and nicer. > + less buggy. > As to me , the choice is obvious. > > and when we implement a proper 64-bit system with e.g. 61-bit SmallIntegers it will no longer be 3 times slower :) Sorry for the derail, but why not 62? ;) http://forum.world.st/Fwd-Plan-discussion-communication-around-new-object-f… Cheers, Henry

April 17, 2013

Re: [Pharo-project] Not all projects on bugs.pharo.org

by Marcus Denker

On Apr 17, 2013, at 11:23 AM, Damien Cassou <damien.cassou(a)gmail.com> wrote: > I don't see the "Ubuntu packaging (ppa:cassou/pharo)" fogbugz project > on bugs.pharo.org > Yes, this seems to be not updated.. I added a bug report a while ago: https://pharo.fogbugz.com/f/cases/10157/http-bugs-pharo-org-is-not-in-sync-… Marcus

April 17, 2013

Re: [Pharo-project] do you know what is this?

by kilon

Estaban correct me if I am wrong but AFAIK MACOS GUI is based on opengl and it is hardware accelerated. I hear the same applied for Linux GUIs too in large. Anyway, my only objection is the fact that we will have twice the code to maintain and more to worry about. As I said I am on macos , I am a mac developer (not pro) still ObjC never made sense for me since it locks me down to a specific platform. It does however make sense to have platform specific code for providing better and smoother support for that platform. So its definetly good to have and thank you for your efforts. Speedwise , I know the subject on optimisation is a huge one and there thousands of ways to speed up specific code. Its a more a practical one, since not to many people work on athens we need to choose one way that makes life easier for everyone and especially people who develop Athens. Part of the sucess of Cairo is that is an easy to understand little library that is very well documented. I would like to see Athens maintain that. -- View this message in context: http://forum.world.st/Fwd-do-you-know-what-is-this-tp4681962p4682078.html Sent from the Pharo Smalltalk mailing list archive at Nabble.com.

April 17, 2013

[Pharo-project] Not all projects on bugs.pharo.org

by Damien Cassou

I don't see the "Ubuntu packaging (ppa:cassou/pharo)" fogbugz project on bugs.pharo.org -- Damien Cassou http://damiencassou.seasidehosting.st "Success is the ability to go from one failure to another without losing enthusiasm." Winston Churchill

April 17, 2013