@mcbone

i have just set it up and it run bouncing ball demo

_________________
maybe i am dyslexia

An Apple a day keep bill gates away

@clebin

Yeah, I wish a little more work had been put into MIST video output - it's VGA is so "hard" it burns long lasting marks on the LCD monitors I have used. Meaning I can see the screen bar and read "Workbench Screen" as well as icons and stuff, hours after I have unplugged the MIST and the monitor is using Display port on a Linux PC!

https://kolla.egnyte.com/dl/T4nk0XT7sj

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

@wawa

Quote:

it has always been an argument in favour of fpga arcade, that it will have an edge over mist, potentially.. but weeks and years pass and so far even technically it seems that mist has an edge over fpgaarcade, especially in therms of speed,

I hear what you're saying. Mike is a brilliant engineer (plenty of the cores on Mist originate from his code) but his time estimates have been unrealistic. I have both on my desk and although the Mist's core is faster (currently) I prefer using the FPGA Arcade because of the video-output. That's a personal preference though.

Quote:

and also for number of other reasons, like that the core is open as far as i know, also mist seems a more pragmatic approach.

This has been sorted now. There's a public SVN server where the code is. He is true to his word, but again his timescales need to be taken with more than a pinch of salt. I don't know how many machines he's managed to ship now, but that's always been the biggest problem IMO and the reason why more developers haven't jumped on board.

Quote:

it has always been my opinion that fpga accelerators for genuine hardware should come first. thinkering with amiga chipset reimplementation is fine and may be lot of fun but it can take years to improve as we have been witessing. it should be treated as open work in progress subsequent to hardware release, because there is even a remaining number of people who would contribute.

I disagree. These are good devices and totally usable already. The MiST has only just got AGA, but it was a good OCS/ECS machine for long before that. I'm very excited about the Vampire 500, but I don't see it as an either/or situation.

Chris

@kolla

Ooh, that's not good!

Chris

@wawa

Quote:

wawa wrote: it has always been my opinion that fpga accelerators for genuine hardware should come first.

The free cores available have a speed less than 68030/25MHz, the Apollo cores might be good but none of them is available currently.


Quote:

wawa wrote: thinkering with amiga chipset reimplementation is fine and may be lot of fun but it can take years to improve as we have been witessing.

I have seen a lack of interest on potential users, the support of hardware manufacturers is hardly existent. If Dennis van Wheren had not made it for its own purpose and gave it to the public for free, there would have never been a Minimig.


Quote:

wawa wrote: it should be treated as open work in progress ...

Currently you can get a free Minimig Core with AGA, but exsiting FPGA accellerators could hardly compete with CPUs that are still in production.

Last edited by OneTimer1 on 17-Sep-2015 at 03:59 PM.

@clebin

Quote:

I disagree. These are good devices and totally usable already. The MiST has only just got AGA, but it was a good OCS/ECS machine for long before that. I'm very excited about the Vampire 500, but I don't see it as an either/or situation. Chris

maybe ia have not made myself clear. i dont see it as either/or, nor im saying thet the chipset cores are not good enough. im rather reffering to the natami-kind of situation, where the consequence would be first to concentrate on a product that satisfies an existing demand, namely for accelarators to move later to standalone systems, as apollo team apparently has acknowledged in the end.

now, even though im not in need of hardware replacement, but im sorry to hear vga output of mist is unsatisfactory. it a hardware issue i take it?

@OneTimer1

Quote:

The free cores available have a speed less than 68030/25MHz, the Apollo cores might be good but none of them is available currently.

i think appollo core with 060 range of performance have been freely available for vampire users, am i wrong? few days ago we have been presented benchmarks with implemented fpu. other than that would dragonfire outperform 030/25hz? i doubt..

Quote:

I have seen a lack of interest on potential users, the support of hardware manufacturers is hardly existent.

who are the manufacturers you have in mind? so far i think support by the ones behind existing amiga fpga projects lokks like it isnt maybe totally professional, but still existant to a good estent. one have to keep in mind these are hobby projects, and have to be considered as such. when it comes to the few actual companies in question, there is no support as long as there is no product, but this is their own choice to be made. if they see no business in it, so be it. i would rather see more open amateur projects rather than any noth thought out products, announced and then delayed for years.

@wawa

Quote:

now, even though im not in need of hardware replacement, but im sorry to hear vga output of mist is unsatisfactory. it a hardware issue i take it?

i wouldn't call it unsatisfactory exactly. On my setup, the FPGA Arcade is like an Amiga with a flicker-fixer, whereas the Mist is more like standard Amiga output. It's just that once you get used to a crisp, rock-solid display with an Indivision AGA or whatever, it's hard to go back to the standard output.

With the Mist I get feint vertical bands similar to what you sometimes see on an A500. Part of the issue might be my monitor - it attempts to scale the display and makes a hash of it, meaning some pixels are anti-aliased.

On the other hand, there's a whole thread on the Mist forum about using a Mist with various hardware upscalers. I guess with the right VGA upscaler, you could get a great picture out of the Mist. It might also help prevent the issue that Kolla's having too.

Chris

Last edited by clebin on 17-Sep-2015 at 06:02 PM.

@wawa

Quote:

wawa wrote: @OneTimer1 Quote: The free cores available have a speed less than 68030/25MHz, the Apollo cores might be good but none of them is available currently. i think appollo core with 060 range of performance have been freely available for vampire users, am i wrong?

Yes you are wrong:
1st.
The Apollo core is not free, even if some people have an agreement for usage, Gunnar announced it will be a commercial product.

2nd.
The 60(?) existing Vampires are too small for the Apollo core and the Vampire2 is not available and suitable only for a reduced Apollo core.

You can read this here:
http://amiga-news.de/de/news/AN-2015-08-00044-DE.html

and you will find some details in the thread.

Quote:

Quote: I have seen a lack of interest on potential users, the support of hardware manufacturers is hardly existent. who are the manufacturers you have in mind?

People like Jens Schoenfeld for example, at least one of his product is using a Minimig core.

Last edited by OneTimer1 on 17-Sep-2015 at 07:21 PM.

@OneTimer1

Quote:

Yes you are wrong: 1st. The Apollo core is not free, even if some people have an agreement for usage, Gunnar announced it will be a commercial product.

it is apparently intended to remain closed source andeventually become commercial. while id prefer it to be opnen i fully understand when people want some return of their apparently quite exceptional work. so far the core seems to be available for vampire1 users for free, which is a bonus in my eyes, nothing to be taken for granted. and despite its limitations due to fpga choice it has been powerful enough to prove them right. if they drop vampire1 support it is understandable given additional work needed to squeeze the core in it. the cards have been cheap, and remain within their limits as functional as they have ever been.

Quote:

Vampire2 is not available and suitable only for a reduced Apollo core.

is that so?
http://eab.abime.net/showthread.php?p=1041172&highlight=vampire+apollo#post1041172

quoting kipper last post from few days ago:

Quote:

Today is a good day for the project, Beta boards are in the hands of the devs and bug squishing is commencing, another bug squished this morning so its getting there Work on adding the FPU to the core is now starting (it is there, needs polishing), there is lots of room left in the FPGA for lots more enhancements that are already being worked on. Workbench is obviously bootingotherwise there wouldnt be no screenshots, so very good signs. the test vampire has 32bit memory, only 64mb of mem, and lower mhz to assist in the bug squishing and the final card is 64bit and 128mb... start your emulators (k, the emulator part is for me

Last edited by wawa on 17-Sep-2015 at 07:36 PM.

@wawa

Quote:

wawa wrote: Quote: Vampire2 is not available and suitable only for a reduced Apollo core. is that so?

Yes, according to the facts posted on my and your link, its not available.

Quote:

..., Beta boards are in the hands of the devs ... ... the test vampire has 32bit memory, ... the final card is 64bit ...

Look closely to the facts you posted: There is/are beta board(s) which differs from the final card, that is very contrary to a card that is available.

And the 64 bit looks like if the Vampire2 will be cancelled in favour for a new design, this means another 3-9 month until it could be sold.

Last edited by OneTimer1 on 17-Sep-2015 at 08:20 PM.

@OneTimer1

Quote:

And the 64 bit looks like if the Vampire2 will be cancelled in favour for a new design, this means another 3-9 month until it could be sold.

im not sure. perhaps the design is the same only the parts differ, or the memory controller uses currently just half of the bus width or something like that. i seriouly doubt the final cards are going to be another design.

Last edited by wawa on 17-Sep-2015 at 08:23 PM.

@Rob

Quote:

Rob wrote: @OneTimer1 It's mentioned in A-EON's press release from April. http://www.a-eon.com/PDF/News_Release_Summit.pdf

That's old. There's no new after that.

Anyway, and like I stated, another 68000-based Minimig doesn't make sense.


@olegil

Quote:

olegil wrote: @cdimauro You don't see your own pattern of jumping into ANY thread with ANY topic whatsoever and promoting Intel products? Seriously?

If you know of an HDMI dongle in the range of $150 (or even less!), with similar specs and computing power, very good to run (Win)UAE, I've no problem reporting it as an example. I simply know only the solution which I've already reported.

Just to be clear: the question was and is technical, and isn't bounded to a single brand.

Any suggestion? We can build an interesting list that can be useful to the other amigans too.
Quote:

On the technical side you're spot on, though. It would have needed something else than 68000. I would have thought this would be the perfect match for one of Jens' accelerator boards.

Exactly. A 68000 is a too old and too simple solution, if you want to embrace the vast majority of the Amiga's software base, and I don't talk only of games.


@OneTimer1

Quote:

OneTimer1 wrote: @cdimauro Quote: cdimauro wrote: The 68000 processor on the Minimig+ is faster than ... The Minimig+ was announced using a Dragonball processor that's more like a 68030 with 30Mhz clock.

It should run at a very high clock, since a 68000 is much slower than a 68030, clock for clock.
Quote:

Quote: cdimauro wrote: A hardware board makes sense only if ... This point might kill every new Amiga hardware.

It can be, but you have to offer something more, otherwise it isn't convenient, or even doesn't make sense, to invest money on a product.

I think that we have to coldly examine the situation. What an amigan need is revival more or less the Amiga experience. You're on front of the monitor (or TV) with a keyboard, a mouse, a joystick (I hate joypads!!!). You select a game and start enjoying the same graphic and audio in the same way that made you happy in the past. Or let the Workbench load, and double clicking an icon you see and interact again with your favorite application.

It doesn't matter if all that stuff is reproduced through an hardware board, a PC, a tablet, a smartphone, etc.. What's important is that you have again the same things, the same feelings.

As I repeated several times: y = f(x). I don't care about "f": I care about "y", with my input "x". And to get the "y", you have to make a deal with your wallet.

That's all.

@wawa

Quote:

wawa wrote: @OneTimer1 Quote: And the 64 bit looks like if the Vampire2 will be cancelled in favour for a new design, this means another 3-9 month until it could be sold. im not sure. perhaps the design is the same only the parts differ, or the memory controller uses currently just half of the bus width or something like that.

They will need a new board, you can't have a 32bit memory and replace it with a 64bit memory without doubling the data lines.

Fast CPUs needs cache and burst memory access, cache needs more chip area than all the other parts on a CPU, this is one of the main reasons why all the free 68k cores could not beat a real 68030.

@cdimauro

Quote:

That's old. There's no new after that.

4 months means old? You live fast life.

@cdimauro

Dude, the _only_ problem with 68000 is Amiga developers through the times not having expected fast 68000 implementations, and insisting on only supporting 020+. I cannot grasp why OS3.5 and 3.9 did not have 68000 support, it would have had no extra cost, the very few corner cases where it would have made difference, two binaries could have been compiled.

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

@pavlor

Quote:

pavlor wrote: @cdimauro Quote: That's old. There's no new after that. 4 months means old? You live fast life.

More or less. I think that software evolves much much faster than hardware.

@kolla

Quote:

kolla wrote: @cdimauro Dude, the _only_ problem with 68000 is Amiga developers through the times not having expected fast 68000 implementations, and insisting on only supporting 020+. I cannot grasp why OS3.5 and 3.9 did not have 68000 support, it would have had no extra cost, the very few corner cases where it would have made difference, two binaries could have been compiled.

I don't remember if version 3.0 supported only 68020. However it made sense to use it as a basis, since all AGA machines had at least a 68020. That was the point, and I agree with the decision they made.

Look: we are talking about late 1992, not 1985. After 7 years continuing to stick with a 68000 is too much. IMO.

@cdimauro

However, in HINDSIGHT, it would have been more convenient NOW if it was possible to just clock a '000 implementation extremely fast rather than having to add all the extras from the '020+.

But this is obviously not part of the original design goals.

Btw, you can run 3.1 on 68000.

I still think any new HW recreations of amiga chipsets should be compatible with existing accelerator cards.

_________________
This weeks pet peeve:
Using "voltage" instead of "potential", which leads to inventing new words like "amperage" instead of "current" (I, measured in A) or possible "charge" (amperehours, Ah or Coulomb, C). Sometimes I don't even know what people mean.

@olegil

Quote:

I still think any new HW recreations of amiga chipsets should be compatible with existing accelerator cards.

what do you mean by compatible to accelerator cards? they should be compatible to 68k instruction set, which as i have been told actually brakes down to be 020 compatible, as it has the widest instruction set of all implementations, others being mostly a subset of it.

@cdimauro

Quote:

cdimauro wrote: I don't remember if version 3.0 supported only 68020. However it made sense to use it as a basis, since all AGA machines had at least a 68020. That was the point, and I agree with the decision they made. Look: we are talking about late 1992, not 1985. After 7 years continuing to stick with a 68000 is too much. IMO.

No, we are not talking of late 1992 - we are talking year 1999, when OS3.5 was released by Haage&Partner, and 2000 when OS3.9 was release.

All CBM incarnations of Amiga OS run fine on 68000. Before 3.1 was released, many friends of mine soft kicked A500 with 3.0 roms ripped from A1200 and installed 3.0. No problems. Also, many OS3.0 beta testers (and developers) did not have AGA systems (they were not available) and did their testing (and development) on systems with only 68000.

The requirement of 68020 was some bright idea that came about when Haage&Partner did OS3.5, and 3.9. I have gone through all components of OS3.9 and the following boingbag packs to pick out the bits and parts that really require 68020 - they are rare and far between! - very little in OS3.9 is "optimized" for 68020, so little that I honestly do not see the point, the requirement is very artificial. My Minimig has a 680SEC00 processor that can run at 50MHz and it is still produced and cheaply available. For 68020 the situation is different. My Minimig runs OS3.9 kickstarts and OS3.9 Workbench. A few important components requires 68020, such as resource.library and I believe parts of Reaction. But Reaction is just ClassAct, for which there is 68000 version that runs fine - another artificial requirement. The biggest annoyance is that OS3.5+ resource.library is 68020+, and without it, the Prefs programs for example will not work. But for most part, one can use 3.1 Prefs programs, the only Prefs program that really changed was the one for Workbench.

Last edited by kolla on 18-Sep-2015 at 02:52 PM.

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC