[Estimated Reading Time: < 1 minute]
A quick post to let everyone know that Allen Bauer kindly answered my direct question in the Embarcadero forums asking for clarification of what “64-bit implementation” means in the XE “Pulsar” slide of the new Delphi roadmap.
The question was ultimately boiled down to this:
Does “64-bit implementation” in Pulsar mean a fully capable and fully integrated Win64 Delphi compiler, debugger and VCL ?
And the answer from Allen was:
Yes. That is the goal. I would say that anything less would be more a “preview” than an actual sellable product.
Which is great news. I am very happy to be wrong on this point.
ahem…
🙂
@Bruce: Yep, well done. You guessed right. However I stand by my original view that the roadmap as initially published was vague and ambiguous on this point and hope/expect to see it updated.
All good assuming ‘Pulsar’ Suddenly doesn’t disappear off a future ‘Roadmap’ or suddenly transform in to a release that doesnt include major features discussed in the ‘roadmap’ a few months before like occured with ‘fulcrumb’ 😀
@FromEurope: There is that too. No matter how unambiguous – or otherwise – the roadmap there is then the question of follow-up.
It is great that we now have a clearer indication that “Pulsar” will be what “Commodore” was supposed to be (and that “Commodore” is what “Fulcrum” was intended to be).
But last time around we had a much clearer indication that Fulcrum would be a cross-platform release (“clearer” in comparison to the current/initial description of Pulsar). What Fulcrum actually ended up containing is a far, far cry from the indication of the clearly stated goal for that release.
@Jolyon,
How is Commodore what Fulcrum was intended to be?
As for Fulcrum being a far cry from what you expected, aside from Mac support being delayed one release, what else is missing? There was talk of a preview Linux compiler early on. Is that what you’re thinking of?
@Bruce: Perhaps I got confused among the plethora of project names, but I thought “Fulcrum” was the (most/more recent) name for the cross platform project. I could be wrong but in this case if I am wrong then so too was Allen Bauer: “we’re working on Fulcrum, the next RAD Studio release with a focus on cross-compilation for Mac and Linux”.
(http://blogs.embarcadero.com/abauer/2010/01/26/38908)
But I have realised that I was a little off… amongst the profusion of project names I missed that the original name for what Commodore is now was in fact “Project X”. The previous goals for “Fulcrum” have been spread across “Pulsar” and “Wheelhouse”.
“Project X” has disappeared from the road-map entirely, or rather “Project X” is now known as “Commodore”.
The headline of “Fulcrum” as previously mentioned was going to be the addition of Mac/Linux cross compilation, but in fact adds neither and doesn’t even incorporate a preview of any such technology.
Trying to straighten all this out is making me dizzy. Stop the roadmap, I want to get off !!!!!!
@Bruce: I’m now even more confused. At one time you yourself seemed to think that “Fulcrum” meant full “Mac and Linux”.
http://stackoverflow.com/questions/2162937/will-the-next-release-of-delphi-be-supporting-cross-compilation-for-mac-and-linux
Take a look at the comments in that question/answer thread. I may be have been wrong with my interpretation of the words on the new roadmap, but I think you will agree that I was spot on with my predictions.
I said: “by the time Delphi 2011 comes around Project X will have been put onto a back burner and some other new and previously unmentioned project being given priority”
Commodore = Project-X renamed and still at the very END of the list.
Pulsar = a previously unmentioned project (unmentioned by that name at least)
I also said: “there really is no reason to believe that Fulcrum will happen either”
And in the sense that what is now being called “Fulcrum” does not in fact contain the big ticket cross-platform items, this too came to pass.
On the other hand, you said: “Fulcrum is the code name for the next release, so yes, the next release will include Mac and Linux support. I don’t know about the time frame, but “this year” seems reasonable. Native 64 bit support will be in a future release. Personally, I prefer this order.”
Whilst Fulcrum was indeed the next release it does not have ANY Mac/Linux support. And “This Year” was indeed the time frame for Fulcrum, but not as you envisioned it and there will be no cross platform delivery until next year (at the very earliest).
We are also now told that we will get Native 64-bit support BEFORE cross compilation.
Oops. 🙂
I would not hold my breath on anything just now. We’ve been “suggested at” that preview compilers would have long since been available.
I would suggest back that sooner or later the big wigs at Embarcadero are going to notice the constant BS and drift and take steps to clean house and get things back on track. Will it work? It seems more likely to just screw things up even more.
I still wish Microsoft had even been allowed into the bidding process. Perhaps if they had one we would have a 64 bit compiler by now, and all this cross platform nonsense wouldn’t be the distraction it is.
I am still waiting for someone to suggest a reasonable explation of why a new linux attempt will be any more successful than the previous attempt. Or perhaps how appropriate it is to expend limited resource chasing after a developer market that serves less than 10% of the total computer market while the OS writer is slowly making it impossible to use any tools other than their own?
I would suggest, based on recent events and decissions, that the days of Delphi suffering from inappropriately focused management are not quite over. Gentlement, fasten your seatbelts – the ride might get bumpy.
Yup. At one point, Fulcrum was going to include 32 bit Mac support and a Linux preview. I was really hoping this meant that full Linux support would be in the following release.
The Linux preview was dropped from Fulcrum early on, but we didn’t know Mac support has being delayed until the new roadmap was released.
Subject to change and all.
If I remember correctly, Commodore (Project X) was originally going to be the first implementation of native 64 bit on Windows. That sounds a lot like what Pulsar is, now, mixed with Mac support delayed from XE.
The Commodore release in this latest roadmap talks about 64 bit cross platform, which is the first “specifics” I remember seeing about this.
Your predictions are like your reading of the roadmap; a little off the mark.
However, you did say that you didn’t think Fulcrum would happen at all. It did happen, but benefit of the doubt suggests that you meant the cross platform features, which did not happen in Fulcrum. Then again, all of your comments seem to be pretty negative, so you were bound to be right on something just by the law of averages. 🙂
Native 64 bit before cross compilation? I don’t understand. I’m pretty sure the Mac and 64 bit support in Pulsar will both use cross compilation. Did I miss a detail somewhere?
@Bruce: Yeah, I’m never right, your’s never wrong. I geddit. 😉
If you have read ALL my posts you will see that I’m only negative when I see things to be negative about. I am positive about the things that I see reason to be positive about. And your law of averages quip applies just as equally to always being positive. Null argument. 🙂
As for the final point…
Originally the order was:
– Cross-compiler with Mac and Linux support but not Win64, together with cross-platform VCL (VCLX) followed later by a Win64 compiler and Win64 version of the current VCL
Now the order is:
– Cross compiler with Win64 support and a Win64 VCL( presumably single sourced with the Win32 VCL), but no cross-platform VCL (VCLX) and no (mention of) any other platform specific VCL’s.
That’s the change of order I meant. I should have qualified “cross compilation” with “cross compilation with a usable cross-platform component library”. But you are usually so good at seeing the intended meaning where it isn’t made explicit… 🙂
In any event, I don’t see any mention of a GUI framework encapsulation for Mac or Linux anywhere on the roadmap…. do you? Without that, and without a cross-platform VCL, just how useful will the compilers for Linux/Mac be ?
Follow-up prediction:
Something in the roadmap will change between now and the time Commodore ships.
Well, you aren’t ALWAYS wrong. And I’m not always positive. You just don’t always see it.
Intended meaning? You must mean reading what was actually written instead of looking for hidden meaning.
I’m pretty sure the Mac support in Pulsar includes the cross platform VCL and RTL. I remember this being the original plan for Fulcrum. Why would you think otherwise (again)? I’m not sure what to expect from server side Linux support, though.
Why does it matter if the GUI framework isn’t in the roadmap? In any event, you’ve probably already heard that it’s going to use QT.
http://delphi.org/2010/02/michael-rozlog/
Do you have a link to the previous roadmap? It looks like the old link was reused for the new roadmap, and I want to refresh my memory.
IMHO the roadmap instead of starting from their intenal codenames should address customers by feature something alike this:
Win64 support: available in Pulsar, new compiler in Commodore
MacOSX support: 32 bit in Pulsar, 64 bit in Commodore
Linux support: 32 bit in whatever, 64 bit in Commodore
etc. etc.
Then
Pulsar: expected Qx 2011
Wheeler: expected Qx 2012
Commodore expected Qx 2013
This way the roadmap would have been much, much clearer
Pulsar = 64bit Delphi WINDOWS and 32 bit MAC OS X (see page 13)
Wheehouse – 64bit C++ (with new compiler backend), Add 32Bit Linux (page 14)
Commodore – everythink in 64bit in new compiler backend and maybe ARM
They don’t wait for new compiler backend for Delphi 64bit.
So do we get a 64bit preview next time or a full capable 64bit compiler and VCL framework? I still don’t get it.
Assuming anything else would be somehow strange.
Guys what we see is a presentation and many implications and dependencies are not discussed as they are understood.
Always consider that some of have little deeper insight than a public presenation allows to show. Mac OSX ist the only “UNIX” OS that currently has “one” windows manager. Under Linux you stell have KDE and Gnome at least … in the end QT or Gnome to consider ….
The way they show makes sense. It is simply let them work. For sure some people who have a demand for Linux maybe go to the lot more common C++ and maybe later can be onboarded … this is something that invevitable even if a Delphi existed or not.
Next Windos and Linux(OSX follow little different 64bit models when we come to C/C++ somehow this can have impact on Delphi at the system level and a cross plattform LLP64 vs. LP64 … as long as you stay in a virtual machine/runtime ….
It’s normally the case that first versions of a major new development (like 64 bit) are buggy. This is just a fact of life but it brings forth howls of complaint every time 🙂
I’m unsure what they mean by “preview” version, but if it means “here it is Delphi Community, give it a try), then that’s good news.
Seems like there was clarification over the weekend. So long as we get a 64bit “everything” windows Delphi environment around this time next year, we won’t pull the trigger on Delphi.
If that slips or the roadmap for Pulsar gets “re-imagineered” into something else, well I think we can all envisage it being a tough roadmap to sell…
On the plus side, if some Mac support is coming in Pulsar with 64 bit Windows, they can’t delay 64 bit windows support until they’ve fixed the Mac support i.e. if the Mac Support or Windows 64 support is flaky in Pulsar, they will have to fix both in parallel and not just fix Mac support and only then deliver their first cut of 64 bit…. I was always concerned about 64bit slipping even further in pursuit of cross platform, but at least this roadmap appears to put the two firmly in parallel and not series.
Gosh, the signal to noise ratio is getting pretty low with all the “he said, she said” stuff …
Can we start talking about what we CAN do with the current and imminent releases 🙂
@David: “She said” ?! Be fair, nobody ever sunk quite *that* low! 😉
@Xepol,
You’re seriously suggesting that Linux serves less than 10% of the overall computer market? Were you born stupid or did Microsoft inject you with some sort of “bullshit serum”?
Linux serves the largest server market! Given that application developers now target the Web industry significantly more than the desktop platform, a lack of a native Linux compiler is suicide for any development toolkit!
Personally I’m not too fussed about native x64 compiling… the advantages there are minimal at best! Being able to provide applications to serve all the major operating systems is ESSENTIAL… not tomorrow, not 5 years from now… RIGHT NOW!
FreePascal has had Linux & Mac compilers for years… it’s had 64bit AND ARM support for YEARS! FreePascal is developed with (at best) a shoe-string budget, so why can’t Embarcadero’s team catch up with all that money behind them?
In my opinion… a lack of common sense, a lack of vision, propensity to focus entirely on the Bottom Line, emphasis always being placed on releasing another pointless “Upgrade” at exorbitant prices every damn 12 months!
@Embarcadero… if you don’t get your heads out of your collective ass and focus on what is important TO YOUR CUSTOMERS instead of TO YOURSELVES, Delphi will die, and you along with it!