forums.silverfrost.com Forum Index forums.silverfrost.com
Welcome to the Silverfrost forums
 
 FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   RegisterRegister 
 ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

32bit good, 64bit bad ? ... A philosophical decision ...

 
Post new topic   Reply to topic    forums.silverfrost.com Forum Index -> General
View previous topic :: View next topic  
Author Message
John-Silver



Joined: 30 Jul 2013
Posts: 1520
Location: Aerospace Valley

PostPosted: Tue Apr 20, 2021 11:21 am    Post subject: 32bit good, 64bit bad ? ... A philosophical decision ... Reply with quote

I may be alone in having noticed:

a) a sudden but inexorable acceleration towards 'only 64bit' updates

b) a (perceived, by me at least) preponderance of bugs relating specifically to 64bit and often not to x32.

here's just a single, simple, example.
http://forums.silverfrost.com/viewtopic.php?t=4456

My question (primarily for the SF developers, but don't let that stop anyone from throwing in their two-penneth of thoughts), a rather philosophical one maybe, is :

why are there so many such posts now with x64 specific bugs ?

what have the (recent) 64bit developments' entailed ? .... a serious re-write of the whole code or just 'tweaking'/minor changes ?

I remember that the 64bit 'convertion' took quite a lot more time than expectd originally, but surely the vast majority of code is the same ?

... or is it just an accumulation of things brought on by a sudden uptake in the usage of 64bit and the detection of such bugs ?

... it's just that I'm having trouble *** this (to me) sudden deluge of bug reports specific to 64bit cobined with what I perciv to be a sudden (and I mean sudden) 'push' to change (update) only for 64bit version.

32bit will still be around for some considerable time yet in the corporate sphere, and i myself believe it's a bit short-sighted to believe that the majority of users will change over to 64bit at the click of one's fingers just because the 32 bit version has stagnated.
It's more likely that a manager will say, when presented with a potential huge bill for a conversion to 64bit, "on yer bike my son !"

Remember what one of our erstwhile fraternity, yes Eddie it's you I'm talking about, has said on many occasions ... that he uses 32bit because of what it can do, stabley and reliably for the most part with workarounds available for the slightly 'wobbly bits' !
I'm sure 95% of users are of the same opinion .... don't try to fix something when it ain't broke !'
Change will come all too soon anyway, and it's never been a straight nor smooth road, so why accelerate down it ? ... which is what seems to be happening.
It appears to be a 'keeping up with the Jones's' philosophy which is under way.
_________________
''Computers (HAL and MARVIN excepted) are incredibly rigid. They question nothing. Especially input data.Human beings are incredibly trusting of computers and don't check input data. Together cocking up even the simplest calculation ... Smile "
Back to top
View user's profile Send private message
Robert



Joined: 29 Nov 2006
Posts: 444
Location: Manchester

PostPosted: Wed Apr 21, 2021 10:23 am    Post subject: Reply with quote

'only 64-bit' updates. Are there many of those?

There have been more 64-bit issues of late and think that is just people taking it up and using it more. There are significant differences between FTN95 in 32 and 64-bit modes. It isn't just a 'recompile' because FTN95 is the recompiler! We hide lots of complexity so that, hopefully, the user can just use a command line switch and move between 32 and 64 bits as required.

There isn't a push for a 64-bit only version. Some features may suit 64-bit more, some might not.
Back to top
View user's profile Send private message Visit poster's website
John-Silver



Joined: 30 Jul 2013
Posts: 1520
Location: Aerospace Valley

PostPosted: Fri Apr 23, 2021 4:58 pm    Post subject: Reply with quote

Robert wrote:
Quote:
'only 64-bit' updates. Are there many of those?

... I've seen enough to give me wobbly legs at the thought of changing something that is stated to be 'only 64bit'only to find out a some later point that it doesn't work for 32 bit !

I definitly get the impression that the 64bit version is 'a bit buggy'.
Hopefully things will iron themselves out over time.

I'm tempted to give a number of %pl a run through with both 32bit and 64bit, but I don't think I've got enough energy to bang my head against a brick wall again after my last foray into the subject which didn't honestly get an awful lot of fixing done. I'm still worried about it's 'robustness'. Or shouldI say lack of it. Maybe 'unpredictable' would be a better term to use.

I saw recently a comment, from Dan I think it was, referring to discussions about 6 months ago concening %pl which I missed completely. Maybe I ought to plough through those if I can find them and it might give me a bit more enthusiasm for 'effort-making'.
_________________
''Computers (HAL and MARVIN excepted) are incredibly rigid. They question nothing. Especially input data.Human beings are incredibly trusting of computers and don't check input data. Together cocking up even the simplest calculation ... Smile "
Back to top
View user's profile Send private message
Kenneth_Smith



Joined: 18 May 2012
Posts: 697
Location: Hamilton, Lanarkshire, Scotland.

PostPosted: Sat Apr 24, 2021 3:00 pm    Post subject: Reply with quote

I use %pl on a daily basis, so it must be robust.
Back to top
View user's profile Send private message Visit poster's website
John-Silver



Joined: 30 Jul 2013
Posts: 1520
Location: Aerospace Valley

PostPosted: Sat Apr 24, 2021 9:38 pm    Post subject: Reply with quote

wasn't that what Julius C. was reputed to say about the coliseum in Rome ... nowlook at it today ! ;Smile

but joking aside, have all the niggly %pl bugs been ironed out now then ?
because 'way back' (a coule of years or so) when I got into using it, it was stated by the powers that be that no further %pl development was on the cards.

numerous 'bugs' relating to the positioning of axes labels (for example) were 'fixed' by recommending the use of silly little parameters to allow the user to 'move' them (by limited amounts).

Re-sizing of a plot (I don't know how many people use that tea & coffee making facility) by dragging the corners, resulted in unpredictable results , sometimes making the plot unuseable when axes labels disappeared off the screen, etc ...

are all those 'bugs' now 'fixed' Paul ?

if not, then your description of %ps as 'robust' might be a little off the mark, for general use, Ken.
_________________
''Computers (HAL and MARVIN excepted) are incredibly rigid. They question nothing. Especially input data.Human beings are incredibly trusting of computers and don't check input data. Together cocking up even the simplest calculation ... Smile "
Back to top
View user's profile Send private message
DanRRight



Joined: 10 Mar 2008
Posts: 2813
Location: South Pole, Antarctica

PostPosted: Tue Apr 27, 2021 2:23 am    Post subject: Reply with quote

John,
Some bugs were fixed in this new incarnation of %pl, some are still there, mostly in LOG part.

But i hope now rest of problems will be fixed quicker. They all are minor anyway. I understand that slow new %pl fixing in the past was because it was no sense to fix some minor things when the entire %pl was scheduled to be completely upgraded.

And of course all settings of the plots may look more graphically visual. There you could, for example, click on X axis numbers instead of searching and scrolling for X axis numbering in the long list of features like today. You can move axis by mouse (like it is done in my own graphics) instead of reading the same long list and clicking 100 times to move it by 100 pixels etcetcetc.

Such Visual %PL will be totally great. Users just need to adopt it ASAP.

Ability to adjusting million things which is done right now with Visual %PL without programming is exactly what needed to satisfy everyone. Because Bill Gates' proverbial 640k will not fit everyone. Before adjusting small misplacement of tics numbers was a nightmare. Today if it exists -- it just a loss of few seconds.

John, i sill assure you to fix this misplacement by few pixels with all 5000 fonts in the computer is special job. It is doable. But it's not justifiable loss of time until many, almost all will start using this new %pl (in my own graphics i gave up and till now did not complete auto adjustment, i also do that manually like Paul is doing with new %pl. Who cares - this is just few clicks).
It probably has no sense to fix it anymore. Paradigm is changed.
Back to top
View user's profile Send private message
Display posts from previous:   
Post new topic   Reply to topic    forums.silverfrost.com Forum Index -> General All times are GMT + 1 Hour
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB © 2001, 2005 phpBB Group