Click Here
home features news forums classifieds faqs links search
6071 members 
Amiga Q&A /  Free for All /  Emulation /  Gaming / (Latest Posts)
Login

Nickname

Password

Lost Password?

Don't have an account yet?
Register now!

Support Amigaworld.net
Your support is needed and is appreciated as Amigaworld.net is primarily dependent upon the support of its users.
Donate

Menu
Main sections
» Home
» Features
» News
» Forums
» Classifieds
» Links
» Downloads
Extras
» OS4 Zone
» IRC Network
» AmigaWorld Radio
» Newsfeed
» Top Members
» Amiga Dealers
Information
» About Us
» FAQs
» Advertise
» Polls
» Terms of Service
» Search

IRC Channel
Server: irc.amigaworld.net
Ports: 1024,5555, 6665-6669
SSL port: 6697
Channel: #Amigaworld
Channel Policy and Guidelines

Who's Online
25 crawler(s) on-line.
 131 guest(s) on-line.
 1 member(s) on-line.


 OlafS25

You are an anonymous user.
Register Now!
 OlafS25:  18 secs ago
 Rob:  5 mins ago
 MEGA_RJ_MICAL:  21 mins ago
 t0lkien:  27 mins ago
 amigakit:  47 mins ago
 OneTimer1:  1 hr ago
 Troels:  2 hrs 2 mins ago
 Gunnar:  2 hrs 17 mins ago
 zipper:  2 hrs 46 mins ago
 NutsAboutAmiga:  3 hrs 51 mins ago

/  Forum Index
   /  Classic Amiga Hardware
      /  News about Vampire and Apollo
Register To Post

Goto page ( Previous Page 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | 32 | 33 | 34 | 35 | 36 | 37 | 38 | 39 | 40 | 41 | 42 | 43 | 44 | 45 | 46 | 47 | 48 | 49 | 50 | 51 | 52 | 53 | 54 | 55 | 56 | 57 | 58 | 59 | 60 | 61 | 62 | 63 | 64 | 65 | 66 | 67 | 68 | 69 | 70 | 71 | 72 | 73 | 74 | 75 | 76 | 77 | 78 | 79 | 80 | 81 | 82 | 83 | 84 | 85 | 86 | 87 | 88 | 89 | 90 | 91 | 92 | 93 | 94 | 95 | 96 | 97 | 98 | 99 | 100 | 101 | 102 | 103 Next Page )
PosterThread
ferrels 
Re: News about Vampire and Apollo
Posted on 23-Jun-2019 15:20:53
#1801 ]
Cult Member
Joined: 20-Oct-2005
Posts: 922
From: Arizona

@NutsAboutAmiga

Quote:
Except you might not find the bug, if you don't have a MMU


I can usually find the bug(s) quite quickly on my cross compiling dev system with its modern debugger.

 Status: Offline
Profile     Report this post  
ferrels 
Re: News about Vampire and Apollo
Posted on 23-Jun-2019 15:25:14
#1802 ]
Cult Member
Joined: 20-Oct-2005
Posts: 922
From: Arizona

@ppcamiga1

Quote:
Last time I use integrated development environment on amiga was 1995. It was Maxon C++. There is no and there will be no IDE with decent compiler on amiga. Nobody will write decent compiler especially for amiga. Not enough users. We will be using gcc to the end.


I agree with you here. This is why I use a cross compiler on my Windows machine. It has a very good IDE and uses GCC which is still being developed. It isn't perfect but it's a far stretch better than using Lattice/SAS C, Aztec or Storm C on real or emulated hardware.

 Status: Offline
Profile     Report this post  
ppcamiga1 
Re: News about Vampire and Apollo
Posted on 23-Jun-2019 16:33:58
#1803 ]
Cult Member
Joined: 23-Aug-2015
Posts: 762
From: Unknown

This whole discussion is boring.
We have year 2019 not 1989.
vampire followers should accept that vampire do not have necesary tools nedeed by developers.
vampire team do not want compatible mmu?
ok.
provide custom tools that can:
- catch null pointer access, and print where in source it is
- something for step by step execution


Last edited by ppcamiga1 on 23-Jun-2019 at 04:34 PM.

 Status: Offline
Profile     Report this post  
Overflow 
Re: News about Vampire and Apollo
Posted on 23-Jun-2019 16:40:30
#1804 ]
Super Member
Joined: 12-Jun-2012
Posts: 1628
From: Norway

@ppcamiga1

sigh..

You were making such nice progress, but had to lapse back.

ferrels has clearly said he thinks the Vampire AND classic alike got lacking development enviroment/tools.

You are constructing conflicts and disagreements where there are none.

Last edited by Overflow on 23-Jun-2019 at 04:41 PM.

 Status: Offline
Profile     Report this post  
ppcamiga1 
Re: News about Vampire and Apollo
Posted on 23-Jun-2019 18:54:36
#1805 ]
Cult Member
Joined: 23-Aug-2015
Posts: 762
From: Unknown

@Overflow

Quote:
ferrels has clearly said he thinks the Vampire AND classic alike got lacking development enviroment/tools.

You are constructing conflicts and disagreements where there are none.


Last answer of ferrels was:

Quote:
I can usually find the bug(s) quite quickly on my cross compiling dev system with its modern debugger.


necesary tools for developers should be made for vampire, not for modern systems, modern systems already have good tools for developers.

 Status: Offline
Profile     Report this post  
Birbo 
Re: News about Vampire and Apollo
Posted on 24-Jun-2019 5:55:06
#1806 ]
Cult Member
Joined: 5-Apr-2007
Posts: 594
From: Zurich, Switzerland

@ppcamiga1

I really, really wonder who you are...

_________________
Sometimes we give people a lot of credit just because they’re writing nice sentences even if it isn’t adding up to much.

 Status: Offline
Profile     Report this post  
thellier 
Re: News about Vampire and Apollo
Posted on 24-Jun-2019 6:26:25
#1807 ]
Regular Member
Joined: 2-Nov-2009
Posts: 263
From: Paris

@Overflow

Quote:
Since Im not a developer, Im asking that you [...] why, when and for what the MMU is needed in development.


The mmu is used for virtual memory
For a given memory adresse it give you the real memory bank or swapped memory
So mainly it is used for giving extra memory

BUT as it knows what memory adresses really exist or not it can give you a clue about when a given program access to non existing memory

So we had tools on 68k like Enforcer (or Grim Reaper on NG) that told you program accessed a bad address in your code in this function with this asm instruction.

[ So if the Apollo team provide an Enforcer tool for 68080 all will be fine ]

So when a program can access a bad memory adresse ?

1) memory not allocated : certainly there is in your code a function that allocate the wanted memory/structure but it may happen that another function use this memory before that

2) memory already freed: certainly there is in your code a function that free memory when it close but it may happen that another function will use AFTER some memory/structure that are no more allocated

3) bad pointer: let give an example
I made a game engine called gmap and there was data describing the levels like tiles,background,sprites,etc...
There was a sprite_num giving the sprite number to use at this place
So I used an array with all the sprites SpritesArray[ ]
And was accessing SpritesArray[sprite_num]
BUT sometime there are no sprite at this place and the game code it as sprite_num=-1

SpritesArray[-1] is not a valid array value nor a valid memory access so bug











 Status: Offline
Profile     Report this post  
Overflow 
Re: News about Vampire and Apollo
Posted on 25-Jun-2019 4:02:08
#1808 ]
Super Member
Joined: 12-Jun-2012
Posts: 1628
From: Norway

@thellier

Thanks for the explaination.

So there are tools that would be very helpful, and features that would speed up development.

Both ppcamiga1 and ferrels agree on this, so its not really a area of disagreement.

Im quite happy to see Stefan Bebbo in active exchange with regards to GCC on Apollo forum;

http://www.apollo-core.com/knowledge.php?b=4¬e=21090

I guess that doesnt really adress the lack of Enforcer or Grim Reaper, but it would be a big step forward to have a actively developed compiler to support Vampire/Apollo spesific features, that would increase exectuable performance.

 Status: Offline
Profile     Report this post  
ferrels 
Re: News about Vampire and Apollo
Posted on 28-Jun-2019 9:15:18
#1809 ]
Cult Member
Joined: 20-Oct-2005
Posts: 922
From: Arizona

@Overflow

Quote:
@thellier Thanks for the explaination. So there are tools that would be very helpful, and features that would speed up development. Both ppcamiga1 and ferrels agree on this, so its not really a area of disagreement. Im quite happy to see Stefan Bebbo in active exchange with regards to GCC on Apollo forum; http://www.apollo-core.com/knowledge.php?b=4¬e=21090 I guess that doesnt really adress the lack of Enforcer or Grim Reaper, but it would be a big step forward to have a actively developed compiler to support Vampire/Apollo spesific features, that would increase exectuable performance.


Where do get off putting words in my mouth that I agree with ppcamiga1? I don't agree with him on much of anything and I have used non-MMU equipped Amigas to debug my code since Amigas were first released. These days I prefer to do my coding AND debugging on a cross-compiling Windows PC because it's much faster than using a real Amiga. That's my preference, but I could just as easily code and debug on a Vampire or classic Amiga even without an MMU, albeit a lot slower. The compilers that I use for coding on classic hardware come equipped with a debugger and an MMU is not required..

 Status: Offline
Profile     Report this post  
megol 
Re: News about Vampire and Apollo
Posted on 28-Jun-2019 12:30:35
#1810 ]
Regular Member
Joined: 17-Mar-2008
Posts: 355
From: Unknown

@ferrels
So you don't think the tools are useful? Because that's what you just denied.

 Status: Offline
Profile     Report this post  
Overflow 
Re: News about Vampire and Apollo
Posted on 28-Jun-2019 13:23:49
#1811 ]
Super Member
Joined: 12-Jun-2012
Posts: 1628
From: Norway

Well, thats my cue to give up on forums.

People just are looking for things to argue about and hair to split.

Last edited by Overflow on 28-Jun-2019 at 01:24 PM.

 Status: Offline
Profile     Report this post  
kolla 
Re: News about Vampire and Apollo
Posted on 28-Jun-2019 19:30:09
#1812 ]
Elite Member
Joined: 20-Aug-2003
Posts: 2859
From: Trondheim, Norway

Bottom line, again (and I am positive that ferrels will spin this into yet another lie about me on the apollo-core forum, poor fellah, so sick and tired of me) is that the Apollo Core has an MMU, and it is used by the core. Anyone who want to do memory tricks will have to go through Gunnar, who may (or may not) implement functionality requested, depending on a number of factors. Why solve in software what can be solved in hardware. That's ok, an open MMU would make the core more attractive outside retro platforms, but that train went long ago now. It just means more work for Gunnar in the long run... forever maintaining and updating the core to fit vampire owners hardware needs, essentially replicating everything that Thomas has done with mmu.library.

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

 Status: Offline
Profile     Report this post  
NutsAboutAmiga 
Re: News about Vampire and Apollo
Posted on 28-Jun-2019 20:15:01
#1813 ]
Elite Member
Joined: 9-Jun-2004
Posts: 12795
From: Norway

@kolla

Reading the specification I can see that 68080 core tries to support self-modifying code. On most modern system changing the code on the fly is considered a unexpected behavior or a bug..

( It's not that you can't change code on PowerPC but you need hazel of flushing instruction cache, and modifying the page so you're allowed write access. )

I wondering if this design decision was done, so that it will not break Amiga500 demos or games.

( I remember it was debated when they created AmigaOS4.0, something about prohibiting programs from make unexpected changes was good idea to prevent old virus from infecting AmigaOS4.x. (It did broke compressed executables, as well.) )

Last edited by NutsAboutAmiga on 28-Jun-2019 at 08:28 PM.
Last edited by NutsAboutAmiga on 28-Jun-2019 at 08:19 PM.
Last edited by NutsAboutAmiga on 28-Jun-2019 at 08:16 PM.

_________________
http://lifeofliveforit.blogspot.no/
Facebook::LiveForIt Software for AmigaOS

 Status: Offline
Profile     Report this post  
kolla 
Re: News about Vampire and Apollo
Posted on 28-Jun-2019 21:36:09
#1814 ]
Elite Member
Joined: 20-Aug-2003
Posts: 2859
From: Trondheim, Norway

Hey ferrels, I see you keep claiming on apollo core forum that I say debugging on Vampire is impossible - which is something I never stated. I see that you also think I have may have some mental issues, yet you are the one obsessing about me stating things I never did.

All I have written regarding MMU is...
* access to an MMU is useful, _also_ for debugging of software, also for Amiga
* access to an MMU is strictly needed for most operating systems
* MMU is also needed for paging to disk or other storage devices, which is handy when dealing with data that are larger than available physical memory, especially with programs that have no concept of disk cache and that rely on all data being in memory - aka just about all Amiga editing software.
* It would be beneficial for Apollo Core to have an accessible MMU, but they chose not to, and instead ride on the claim that noone ever needs it.

You wanted example of when it is more useful to use native tool chain than cross-developing - well, I cannot count the number of times I have fixed endian issues when cross-building for 68k and armeb on x86, not on my software, but in optimistically written build scripts and make files. Sometimes this issue is very subtle and only manifest themselves when running the code. Then there is the problem with toolchains that simply do not exist for x86, code that relies on certain compilers, software with only partly source code availability, and where new components needs to be built using same toolchains as main program to work, etc etc...

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

 Status: Offline
Profile     Report this post  
kolla 
Re: News about Vampire and Apollo
Posted on 28-Jun-2019 21:44:55
#1815 ]
Elite Member
Joined: 20-Aug-2003
Posts: 2859
From: Trondheim, Norway

I forgot to mention one more thing I have written about MMU on Amiga.
* it can be used (and on certain Amiga models is used) to overcome hardware bugs and issues. Especially is this the case for the Zorro3 bus. Gunnar says this is a non-issue, but that is yet to be shown with a Vampire system that has a Zorro3 bus. This does not exist. MMU.library has the benefit of a config file, so the user can overcome various issues by changing a config file. An apollo core based card for A3000/A4000 can quickly need similar approaches to overcome the same issues.

There, ferrels, you can now return to the castle of doom and report about my sanity :)

Last edited by kolla on 28-Jun-2019 at 09:49 PM.

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

 Status: Offline
Profile     Report this post  
khayoz 
Re: News about Vampire and Apollo
Posted on 28-Jun-2019 22:45:36
#1816 ]
Regular Member
Joined: 26-Dec-2004
Posts: 216
From: Stockholm, Sweden

@thread
Interested or not interested in V4, 91 pages..
Evil in League with Satan I have never seen so many negative posts..

Easy fix! You who want it get the F out of this thread!
Trolls starve!

_________________

 Status: Offline
Profile     Report this post  
tygre 
Re: News about Vampire and Apollo
Posted on 29-Jun-2019 3:39:04
#1817 ]
Regular Member
Joined: 23-Mar-2011
Posts: 279
From: Montreal, QC, Canada

@khayoz

+1

_________________
Tygre
Scientific Progress Goes Boing!

 Status: Offline
Profile     Report this post  
Overflow 
Re: News about Vampire and Apollo
Posted on 29-Jun-2019 5:09:37
#1818 ]
Super Member
Joined: 12-Jun-2012
Posts: 1628
From: Norway

@kolla

With regards to the Zorro3 bus issue;

I dont have such hardware, so Im completely unfamiliar with the issue itself.
Until there are cards for said Amiga models, there isnt really a problem, but a expected problem, right?
So IF there actually are plans for cards for these models, THEN its an issue that needs adressing, if the memory managment Apollo hasnt taken it into account yet.

I would have to look for the "roadmap" to see if the team even plans Vampire cards for that branch of Amigas tho. I Tuko said they are focusing on the V4 Standalone. Then I suspect the V1200 is getting some love etc, so I wouldnt expect the A3000/4000 to get any "love" anytime soon, if at all.

So until then, its a hypotetical, altho nice to be aware of ofcourse

I did say I would check out of this thread, but kolla made a spesfic case claim/potential problem, so it would be good if someone with the knowledge to adress it sticks to the spesific topic, instead of attacking his persona

EDIT:

And I asked the question on Apollo Forum, and Gunnar basically confirmed that A3000/4000 is unlikely to get any vampire love.

Quote:
think you are right. The simple reason why, is that making many many different cards is a lot of work. And we will need to work hard to satisfy the A1200 demands.


So any futher discussion regarding this hardware branch is merely academic.

Last edited by Overflow on 29-Jun-2019 at 06:26 AM.
Last edited by Overflow on 29-Jun-2019 at 06:26 AM.
Last edited by Overflow on 29-Jun-2019 at 05:11 AM.

 Status: Offline
Profile     Report this post  
hth313 
Re: News about Vampire and Apollo
Posted on 29-Jun-2019 6:41:38
#1819 ]
Regular Member
Joined: 29-May-2018
Posts: 159
From: Delta, Canada

Whether an MMU is useful or not depends various things. I have used A500 and A2000 without it and I did not have much problem at the time. The debug tools utilizing the MMU were very useful so I got an A3000 mainly for that reason.

This A3000 came with soft kicking version 2.04, which means the machine loaded 2.04 kickstart into RAM and then used the MMU to remap it to the ROM address area and write protect it, so 512K of RAM was turned it into something that basically resembled ROM. On top of this, it could use the MMU based debug tools as well. This was very neat and useful.

Later 3.0 came along and I could just upgrade the machine without buying ROM chips (keep using the MMU in the same way). After the Amiga era, the machine served for some years as a NetBSD box, and that one has memory protection and virtual memory, making good use of the MMU. In fact, it could not have used NetBSD without the MMU.

Today, the machine is still soft-kicking, but now it loads two 512K AROS ROM in a similar way into RAM and remap and write protect them to mimic ROMs. The images used are the genuine AROS ROMs compiled for the Amiga ROM address areas.

Most people booting AROS do so on top of AmigaOS, but using a relocatable ROM image file. This can be loaded to any address. Machines that have MMU can AFAIK write protect it too in a similar way, others run it without write protection as far as I understand (as there is no MMU on many machines).

My machine has used the MMU from day 1 and have used it for various purposes. Having said that, many Amiga users can probably live happily without it. We can have different opinions on whether it is needed by developers. Most developers I have met tend to like whatever assistance they can get on finding problems. Having said that, I do not use the MMU at the moment for debugging, but I am only a casual Amiga developer these days.

What the Vampire team is up to here I do not know. Maybe they allow for flash based ROMs, which would limit the need for an MMU. Running Unix on it is probably not so much of a point today. Any RPi or one of its cousins will probably do that better at a small cost. Some few developers may have some use for an MMU and some other use-cases may perhaps pop up, and perhaps it is not so important. On the other hand, AmigaOS did not really have need for it and yet my A3000 has gone through lots of creative uses of its MMU, as it happened to be there and could be used to solve various problems that would not have been possible (or as easy) otherwise.

 Status: Offline
Profile     Report this post  
ferrels 
Re: News about Vampire and Apollo
Posted on 29-Jun-2019 6:47:36
#1820 ]
Cult Member
Joined: 20-Oct-2005
Posts: 922
From: Arizona

@megol

Quote:
@ferrels So you don't think the tools are useful? Because that's what you just denied.


Not so much for people using cross-compilers on a Windows or Linux box....

Last edited by ferrels on 29-Jun-2019 at 06:48 AM.

 Status: Offline
Profile     Report this post  
Goto page ( Previous Page 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | 32 | 33 | 34 | 35 | 36 | 37 | 38 | 39 | 40 | 41 | 42 | 43 | 44 | 45 | 46 | 47 | 48 | 49 | 50 | 51 | 52 | 53 | 54 | 55 | 56 | 57 | 58 | 59 | 60 | 61 | 62 | 63 | 64 | 65 | 66 | 67 | 68 | 69 | 70 | 71 | 72 | 73 | 74 | 75 | 76 | 77 | 78 | 79 | 80 | 81 | 82 | 83 | 84 | 85 | 86 | 87 | 88 | 89 | 90 | 91 | 92 | 93 | 94 | 95 | 96 | 97 | 98 | 99 | 100 | 101 | 102 | 103 Next Page )

[ home ][ about us ][ privacy ] [ forums ][ classifieds ] [ links ][ news archive ] [ link to us ][ user account ]
Copyright (C) 2000 - 2019 Amigaworld.net.
Amigaworld.net was originally founded by David Doyle