Why no new accelerators?
- Thread starter mjnurney
- Start date
- Replies 188
- Views 2200
The bulk of those thread views will significantly pre-date the posting of the poll.
And most people wont even bother to vote leaving it to someone else. At least your ratio of no and yes is positive 
But I agree with Justin in the MOST realistic way to keep Amiga going. Port to X86.
Another option would be to try and edge into the ARM mobile market. Amiga OS is trim and fast enough to run like awesomeness on a 1GHz Snapdragon or what have you. Go mobile/tablet, that's the future anyway. Go after Android and iOS. Its almost ridiculous to see Amiga games being ported to iOS and Andriod. Amiga OS should be there alongside these running them natively!
I'd love an Amiga powered Phone and Tablet!
I'm chuffed to bits that a Netbook is coming at least, but its still gonna be powered by an ancient PPC. Better than a smack in the face though!
But I agree with Justin in the MOST realistic way to keep Amiga going. Port to X86.
Another option would be to try and edge into the ARM mobile market. Amiga OS is trim and fast enough to run like awesomeness on a 1GHz Snapdragon or what have you. Go mobile/tablet, that's the future anyway. Go after Android and iOS. Its almost ridiculous to see Amiga games being ported to iOS and Andriod. Amiga OS should be there alongside these running them natively!
I'd love an Amiga powered Phone and Tablet!
I'm chuffed to bits that a Netbook is coming at least, but its still gonna be powered by an ancient PPC. Better than a smack in the face though!
Justin
Active member
you know what? i think that is actually a sensible idea!! mobile os would work really well, i almost think that is what the amiga os was made for
And most people wont even bother to vote leaving it to someone else. At least your ratio of no and yes is positive
But I agree with Justin in the MOST realistic way to keep Amiga going. Port to X86.
Another option would be to try and edge into the ARM mobile market. Amiga OS is trim and fast enough to run like awesomeness on a 1GHz Snapdragon or what have you. Go mobile/tablet, that's the future anyway. Go after Android and iOS. Its almost ridiculous to see Amiga games being ported to iOS and Andriod. Amiga OS should be there alongside these running them natively!
I'd love an Amiga powered Phone and Tablet!
I'm afraid I agree with Stan's sentiments
To me there's not much to be gained in this sort of thing. Sure, it'd be great but what could you run on it? Could you use it for your day to day computer?
No.
After 400 people had enjoyed putting things together, sorting issues, getting some semblance of stability in the system, the next big cry would be "WHERE'S THE SOFTWARE", headed by the perennial call for a decent browser for AOS3.
Then there'd be the call for better graphics. And better IO, whether disk or USB. And ultimately all your Amiga is doing is being a host to something else, and routing keyboard and mouse inputs to the new hardware. A by-product would be crap reliability as 25 year old capacitors, passives and ICs would be feeding new stuff...
It's an interesting challenge, for sure, but I think the sweet spot has been found by Jens, and not merely by chance or accident.
The majority of software was designed for lightly accelerated machines. Not a lot of new software exists that builds on this in classic AOS3 / 68k. Yes, there's some interesting demos, but...
Future? If we can keep this stuff running, an 030 or 040 is fine for the classics. Or perhaps a replacement motherboard that is x86 based, or with real Amiga chips, or FPGA amiga chips, and interfaces with the 500 and 1200 keyboard... Or NatAmi. Or FPGAArcade.
Beyond that, AROS is Amiga-like and once AOS4 gets over its obsession with non-x86 it might be interesting.
I love the classics, and I use them: sitting next to me are my GBA1000, 3000 and 4000 and they're connected to the same monitor that the computer I'm posting this from is connected to.
But a shot in the arm, cpu-wise, will just highlight the other areas of weakness (software, os, graphics, sound) and leave us thinking, "that's cool, but now what"?
---------- Post added at 11:14 ---------- Previous post was at 11:12 ----------
AOS on Raspberry Pi... Damn it, that's hugely tempting! A simple, lightweight OS on a simple, lightweight machine.
Problem is that AOS doesn't give enough of what is expected from an OS in the V3 incarnation. AOS4 on ARM... Ohhh yesssss....
Put an emulation layer on there which mapped $dff180 to background colour etc and you'd have a great learner system...
To me there's not much to be gained in this sort of thing. Sure, it'd be great but what could you run on it? Could you use it for your day to day computer?
No.
After 400 people had enjoyed putting things together, sorting issues, getting some semblance of stability in the system, the next big cry would be "WHERE'S THE SOFTWARE", headed by the perennial call for a decent browser for AOS3.
Then there'd be the call for better graphics. And better IO, whether disk or USB. And ultimately all your Amiga is doing is being a host to something else, and routing keyboard and mouse inputs to the new hardware. A by-product would be crap reliability as 25 year old capacitors, passives and ICs would be feeding new stuff...
It's an interesting challenge, for sure, but I think the sweet spot has been found by Jens, and not merely by chance or accident.
The majority of software was designed for lightly accelerated machines. Not a lot of new software exists that builds on this in classic AOS3 / 68k. Yes, there's some interesting demos, but...
Future? If we can keep this stuff running, an 030 or 040 is fine for the classics. Or perhaps a replacement motherboard that is x86 based, or with real Amiga chips, or FPGA amiga chips, and interfaces with the 500 and 1200 keyboard... Or NatAmi. Or FPGAArcade.
Beyond that, AROS is Amiga-like and once AOS4 gets over its obsession with non-x86 it might be interesting.
I love the classics, and I use them: sitting next to me are my GBA1000, 3000 and 4000 and they're connected to the same monitor that the computer I'm posting this from is connected to.
But a shot in the arm, cpu-wise, will just highlight the other areas of weakness (software, os, graphics, sound) and leave us thinking, "that's cool, but now what"?
---------- Post added at 11:14 ---------- Previous post was at 11:12 ----------
AOS on Raspberry Pi... Damn it, that's hugely tempting! A simple, lightweight OS on a simple, lightweight machine.
Problem is that AOS doesn't give enough of what is expected from an OS in the V3 incarnation. AOS4 on ARM... Ohhh yesssss....
Put an emulation layer on there which mapped $dff180 to background colour etc and you'd have a great learner system...
Summary:
As has been said, the FPGA market is not at the moment financially viable for low quantity production. It will be several years before this does become viable. x1000 is for many outwith their budget and still does not resolve the issue of maintaining the niche retro 68k market and production of new accelerators, which brings us back to:
Why no new accelerators?
Financial viability: The risk of producing new accelerators is too high to guarantee a return when the market is a niche market and demand is variable.
The market demand and price must absorb the Development and production costs.
Based on the price that people are willing to pay for 030 second hand cards which tends to max out at £170, averaging around the £150 mark. This would lead me to think that the max price ranges for this 060 card would be from £150 to £250 depending on demand.
To determine if this is feasible at all you would need to poll what people would be willing to pay. This would be the basis of a feasibility study comparing the costs of components and manufacturing and the return on investment.
Availability: The risk of of financial loss is too high to produce significant quantities.
A minimum preorder/prepaid quantity would have to be met to reduce risk and begin production.
To generate interest a target spec and price would have to be determined.
On a practical level this is the spec that has been suggested.
Updated A1200 expansion card
060 processor (internal MMU)
Supplied with 256Mb memory onboard
FPU, RTC(Optional)
Basically an updated ACA card
Questions:
Could AmigaKit negotiate with Jens to produce such a card? Jens already has tooling for the connectors so an investment in tooling from AmigaKit would sweeten the deal.
What is the minimum preorder/prepaid quantity required before a batch of could be produced. Of course this would depend on a price that would attract the market and the demand on that price.
Comments have been made on four other points
Internal Port card(s) to support modern tech
USB 2 ports (+powered)
SATA I/II ports + IDE bridge
10/100/1000 ethernet port
Digital Audio
RTG Graphics (+Legacy AGA)
Obviously this requires other projects to be established or wait for Jens to do them.
User hacking of mainboards.
The issue expressed was that many users are unwilling or incapable of hacking their own mainboards.
To resolve the issue a dedicated service center would have to be available from AmigaKit and possible other commercial retailers. They could employ the services of some of the more experienced hackers on this and other sites to do the fixes. Pocket money for the service engineers?
User hacking of case.
The issue expressed was that many users are unwilling or incapable to hack their own cases.
Since the case is owned and patented and the cost of licensing and production is most likely not financially viable, the only other alternative is a modding service provided by others as was described for hacking mainboards.
Why bother?
If there is a market for such hardware and it is financially viable then there is no reason not to go into production to maintain the retro side of things.
As has been said, the FPGA market is not at the moment financially viable for low quantity production. It will be several years before this does become viable. x1000 is for many outwith their budget and still does not resolve the issue of maintaining the niche retro 68k market and production of new accelerators, which brings us back to:
Why no new accelerators?
Financial viability: The risk of producing new accelerators is too high to guarantee a return when the market is a niche market and demand is variable.
The market demand and price must absorb the Development and production costs.
Based on the price that people are willing to pay for 030 second hand cards which tends to max out at £170, averaging around the £150 mark. This would lead me to think that the max price ranges for this 060 card would be from £150 to £250 depending on demand.
To determine if this is feasible at all you would need to poll what people would be willing to pay. This would be the basis of a feasibility study comparing the costs of components and manufacturing and the return on investment.
Availability: The risk of of financial loss is too high to produce significant quantities.
A minimum preorder/prepaid quantity would have to be met to reduce risk and begin production.
To generate interest a target spec and price would have to be determined.
On a practical level this is the spec that has been suggested.
Updated A1200 expansion card
060 processor (internal MMU)
Supplied with 256Mb memory onboard
FPU, RTC(Optional)
Basically an updated ACA card
Questions:
Could AmigaKit negotiate with Jens to produce such a card? Jens already has tooling for the connectors so an investment in tooling from AmigaKit would sweeten the deal.
What is the minimum preorder/prepaid quantity required before a batch of could be produced. Of course this would depend on a price that would attract the market and the demand on that price.
Comments have been made on four other points
Internal Port card(s) to support modern tech
USB 2 ports (+powered)
SATA I/II ports + IDE bridge
10/100/1000 ethernet port
Digital Audio
RTG Graphics (+Legacy AGA)
Obviously this requires other projects to be established or wait for Jens to do them.
User hacking of mainboards.
The issue expressed was that many users are unwilling or incapable of hacking their own mainboards.
To resolve the issue a dedicated service center would have to be available from AmigaKit and possible other commercial retailers. They could employ the services of some of the more experienced hackers on this and other sites to do the fixes. Pocket money for the service engineers?
User hacking of case.
The issue expressed was that many users are unwilling or incapable to hack their own cases.
Since the case is owned and patented and the cost of licensing and production is most likely not financially viable, the only other alternative is a modding service provided by others as was described for hacking mainboards.
Why bother?
If there is a market for such hardware and it is financially viable then there is no reason not to go into production to maintain the retro side of things.
I`ll tell you where the software is and its AROS, Toni and co have done great things recently with this and it is now able to boot on an A1200 or A4000 but I am thinking that with heaps of memory and a faster IO and CPU it be a whole lot better. oh and thanks for the free kickstart replacement Toni
Is there any chance we can leave off with the there`s no point to this or it will never happen posts, cos really if everyone says it not going to happen then it probably won`t and like wise the more we talk about it more chance we have of seeing something new happen
all I want is more RAM, A faster CPU, and Better Storage Options. Theres nothing to say there can`t be a proper expansion bus on there as well with a connector thats everywhere and is`t going to disappear in 10 years time
Is there any chance we can leave off with the there`s no point to this or it will never happen posts, cos really if everyone says it not going to happen then it probably won`t and like wise the more we talk about it more chance we have of seeing something new happen
all I want is more RAM, A faster CPU, and Better Storage Options. Theres nothing to say there can`t be a proper expansion bus on there as well with a connector thats everywhere and is`t going to disappear in 10 years time
I agree - Aros is the software, but it doesn't need classic Amiga hardware to run on, which is why I sort of think it's best off leaving classics to the 030/040 generation of processors.
So... You want AROS, faster CPU, more RAM, and better storage options... Serious question, not trolling: why does that need to be on Amiga hardware?
So... You want AROS, faster CPU, more RAM, and better storage options... Serious question, not trolling: why does that need to be on Amiga hardware?
Any new Amiga OS should just transfer to current pc components and use WinUAE to run 68k software. Embrace today's hardware with a well written efficient OS like it always was or leave things in the past. Not to mention some software companies may even support an OS which has the potential to sell and establish some sort of decent sized user base, but running a new OS on non-standard or old hardware is just madness.
Why is simple because its an Amiga and we still can do it, sure you don`t need an amiga to run AROS but it be way cooler running it on an A1200 or A4000 than some soulless x86 components
obviously I see AROS as a far reaching amiga solution which will give us Architecture that is x86 and 68000 Compatible at the source level, It has a future and is slowly moving along nicely
obviously I see AROS as a far reaching amiga solution which will give us Architecture that is x86 and 68000 Compatible at the source level, It has a future and is slowly moving along nicely
hell,you may as well make a new amiga from scratch if your going to replace the graphics ,the sound ,the ide.
the part that really bothers me about this whole affair is that most peaple here thinks its a bad idea to bolt an 68k cpu to an amiga going on 20 years old,then 2 paragraphs down its wow lets put a ppc in there instead with a gfx card and usb,im sorry but whats the difference,
itll still be plugged into the same machine as the ppc card.
ok fair enough the softwares out of date,so is the machine.
lets face it lads,060's wernt cheap but they where cheaper than ppc no-one on this planets going to be able to make an 060 card from scratch for a few quid more then a low end 030.
and the stuff peaple want on a card you can fit in you hard,my god............and for around 250 quid? ...........and yet its ok to spend 6-700 pounds for a ppc card that yes plugs into the same machine with bad caps?and is only capable of 330mhz then pay what 130 pounds for os4.1
is that cheap? realistic? how much lads to make a new ppc card?
you know its never going to happen because you could get a x86 for less and download winuae..............and you want peaple to work on it for free and hand in there work,sorry lads i just dont get it.
the part that really bothers me about this whole affair is that most peaple here thinks its a bad idea to bolt an 68k cpu to an amiga going on 20 years old,then 2 paragraphs down its wow lets put a ppc in there instead with a gfx card and usb,im sorry but whats the difference,
itll still be plugged into the same machine as the ppc card.
ok fair enough the softwares out of date,so is the machine.
lets face it lads,060's wernt cheap but they where cheaper than ppc no-one on this planets going to be able to make an 060 card from scratch for a few quid more then a low end 030.
and the stuff peaple want on a card you can fit in you hard,my god............and for around 250 quid? ...........and yet its ok to spend 6-700 pounds for a ppc card that yes plugs into the same machine with bad caps?and is only capable of 330mhz then pay what 130 pounds for os4.1
is that cheap? realistic? how much lads to make a new ppc card?
you know its never going to happen because you could get a x86 for less and download winuae..............and you want peaple to work on it for free and hand in there work,sorry lads i just dont get it.
When the 060 add-on card becomes readily available for the FPGA Replay, I will gladly donate the CPU from my Apollo or any other 060 board in my possession should it become necessary.
I don't get what the difference is between a 68k processor's 'soul' and a modern x86's 'soul', or a 68k running in FPGA's 'soul'.
Other than nostalgia. And perhaps the challenge of 'because it's there'. Hell, I've hacked my 1000 as far as it would go 'because I can'
so I completely understand the appeal of that !I wouldn't want to stop anyone building this, absolutely not, but I think what people could do with such things needs to be clear otherwise people will spend a lot of time and a lot of money 'because they can'. In my experience, an 060 is less use for WHDLoad than a decent 030...
If someone's pimping out their Amiga with expectations to do things like surf the web, then memory and a decent processor is great, but the software doesn't exist despite being demanded (often vocally) for 10 years! Then the graphics becomes the bottleneck. Then the mime type support. Then sound.
And yes, I have been down that route...
and turned back from it somewhat!
At somepoint we "may" have a quick 68k comparable machine in the Natami or some other softcore fpga machine with Aga or super aga and these will be the future of the Amiga if and only if production is high and price is low enough to get customers.
As for the fate of the x1000, I can't say but I'm not buying one so I don't care. My original question was why don't we have a new 68k accelerator , preferably an 060 but 040 would do as a min spec. Id personally like to see the Amiga base models dragged out of the slower than a very slow thing that's having an asmatic attack mode.
Why?
Well 030 accelerator a great, I have one.
But
I'd like to play mp3 , play scummvm , run around in doom and maybe , just maybe see quake running on a 25 year old Amiga and yes it is slow on 68k, I know.
I for one, don't give a toss about USB, IDE or Graphics either to be honest. The Amiga is ocs,ecs or aga and that's it folks.
If you add all this stuff, the cost is more than a new Mac book pro and ends up with less software that the zx81.
So for me atleast. Basic 060 or 040 min and maybe 128 / 256 ram if that is possible.
Yes id love LAN,USB,IDE,Sata,super aga, voodoo 5 chipset and an expansion port for my coffee machine (set at 6:55 am) but its not going to happen.
Sorry
As for the fate of the x1000, I can't say but I'm not buying one so I don't care. My original question was why don't we have a new 68k accelerator , preferably an 060 but 040 would do as a min spec. Id personally like to see the Amiga base models dragged out of the slower than a very slow thing that's having an asmatic attack mode.
Why?
Well 030 accelerator a great, I have one.
But
I'd like to play mp3 , play scummvm , run around in doom and maybe , just maybe see quake running on a 25 year old Amiga and yes it is slow on 68k, I know.
I for one, don't give a toss about USB, IDE or Graphics either to be honest. The Amiga is ocs,ecs or aga and that's it folks.
If you add all this stuff, the cost is more than a new Mac book pro and ends up with less software that the zx81.
So for me atleast. Basic 060 or 040 min and maybe 128 / 256 ram if that is possible.
Yes id love LAN,USB,IDE,Sata,super aga, voodoo 5 chipset and an expansion port for my coffee machine (set at 6:55 am) but its not going to happen.
Sorry
Last edited:
Justin
Active member
well you can do all of those things right now if you are willing to spend the money, my A1200 (yes the real A1200) plays Doom at between 26 and 32 fps.
i've yet to try Quake, but i'm pretty sure it'll be unplayable.
on the other hand, £90 and AROS and you can do all the things you listed plus a lot more at crazy (Amiga related) speed and blow any PPC/X1000 out of the water, lol spend £150 and AROS and have a machine that'll kill the X1000 and all of it's later incarnations.
add a catwaesel and you can even play from original Amiga floppies, add a Keyrah and use an original Amiga Keyboard/mouse and joysticks heck it'll even use the original LEDS
the classic amiga emulation is seemles, so you can have all the classics and run "new" (ROFLMAO@new) all in one, and it will be as good as OS4 if not better in some cases.
so we go full circle, Amiga accelerator? why?
i've yet to try Quake, but i'm pretty sure it'll be unplayable.
on the other hand, £90 and AROS and you can do all the things you listed plus a lot more at crazy (Amiga related) speed and blow any PPC/X1000 out of the water, lol spend £150 and AROS and have a machine that'll kill the X1000 and all of it's later incarnations.
add a catwaesel and you can even play from original Amiga floppies, add a Keyrah and use an original Amiga Keyboard/mouse and joysticks heck it'll even use the original LEDS
the classic amiga emulation is seemles, so you can have all the classics and run "new" (ROFLMAO@new) all in one, and it will be as good as OS4 if not better in some cases.
so we go full circle, Amiga accelerator? why?
Justin
Active member
My A1200 is a Commodore ya cheeky monkey
as for the PC Aros thing if Commodore now own the rights to Amiga then in fact the Amiga is now officially a PC:nod:
opcorn:opcorn:
as for the PC Aros thing if Commodore now own the rights to Amiga then in fact the Amiga is now officially a PC:nod:
opcorn:opcorn:
If someone wrote an modern Athlon that booted without windoze and then fitted a small itx board in a 500 case ..
That took the keyboard / mouse & power
Well that's the end of the Amiga classic hardware really.
That took the keyboard / mouse & power
Well that's the end of the Amiga classic hardware really.
here,here,if the lads here want an 030 for whdload thats fair anough,but if peaple want 060 they have to stay focused on what is realistic within the hardware its better that way.
ok,lets put it this way.
when the aca1230 came out it was wow state of the art man,even the model thats 28mhz.yeah its a nice card,if you want whdload on a 4 gig drive.have you seen what dooms like on an 030? i think the peaple saying it was playable becuase of the sdram interface are waering rose coloured glasses if by playable they meant "it loaded and i could play at a slideshow speed"
these same peaple would probably realise that an 060 card has over 10x the cpu power of an 030 card at its "slowest" 50mhz and even more with an fpu.
and then theres 040's there must be loads of peaple that have these in systems ,the fastest 040 at 40mhz is only half the cpu power of an 060 at 50mhz
Similar threads
