|
|---|
Niclas Aronsson
Sweden
| |
Posts 57
24 Feb 2011 16:38
| Thomas Hirsch wrote:
|
Ok, a short update where I am right now:
Unfortunately I found two schematic/layout errors which caused the 1.8V supply to be short circuited. So I had to work around this and this took a little time.
This means I can now safely turn on the board and debug it. Now there are FPGA design changes to do. Namely pinning and IP adaption to the new main FPGA.
|
Nice :) Thanks for the update Thomas.
Will the workarounds be a small patch wire ? or do you plan on making a rev 2 of the board ?
| |
Nathaniel Tolbert
USA
| |
Posts 9
24 Feb 2011 16:50
| I take it that there was a change in the I/O pins available with this new FPGA versus the one used in the previous board? Oh by the way first time poster, long time lurker. Aside from the possibility of more I/O Pins and more Logic Elements on the new FPGA, is there any other major differences between this FPGA and the previous one that was used? Also, when trying to look up information on FPGAs I noticed that the speed rating for most isn't posted. Is there a specific reason for this? I wait patiently as I hope to purchase a Natami when it comes out. Sadly I have no developer skills, so all I can do is consume, as I have no other way to contribute to the project.
| |
Thierry Atheist
Canada
| |
Posts 1828
24 Feb 2011 17:20
| Hi Thomas, I'm guessing, as you didn't actually state it, but, there's no physical damage to the motherboard??? Just the... power supply? Nathaniel Tolbert wrote:
|
I wait patiently as I hope to purchase a Natami when it comes out. Sadly I have no developer skills, so all I can do is consume, as I have no other way to contribute to the project.
|
Hi Nathaniel,Thank you for showing up!!! It's a bold move. We've never met on other message boards, I don't think. You know, even being "just the consumer" helps us A LOT, as many won't climb onto that tree branch at hearing of the NaAmi's price. Think you can't help because of no developer skills? At this point maybe not, but if you have an Amiga, any model except for CDTV or CD32, you can EASILY learn to code in AMOS Professional! And if you aren't able to do that, which I would find being extremely unlikely, you could still later on show other people how fantastic it is to USE a NatAmi after you get one. :-D
| |
Ed Dream
Russia
| |
Posts 28
24 Feb 2011 18:59
| Thomas Hirsch wrote:
|
Hi Nathaniel,
Thank you for
showing up!!! It's a bold move.
We've never met on other message boards, I don't think.
You know, even being "just the consumer" helps us A LOT, as many won't climb onto that tree branch at hearing of the NaAmi's price.
Think you can't help because of no developer skills? At this point maybe not, but if you have an Amiga, any model except for CDTV or CD32, you can EASILY learn to code in AMOS Professional!
And if you aren't able to do that, which I would find being extremely unlikely, you could still later on show other people how fantastic it is to USE a NatAmi after you get one. :-D
|
And about the use .... Many people would like it at the user level to support the project by purchasing a fee for developers. Because in fact we testers on urovsne operating and file systems, debuggers drivers and interfaces.
| |
Ed Dream
Russia
| |
Posts 28
24 Feb 2011 19:01
| Thomas Hirsch wrote:
|
Ok, a short update where I am right now:
Unfortunately I found two schematic/layout errors which caused the 1.8V supply to be short circuited. So I had to work around this and this took a little time.
This means I can now safely turn on the board and debug it. Now there are FPGA design changes to do. Namely pinning and IP adaption to the new main FPGA.
|
That's bad.
Is it possible in this audit, correct? For example changing components. Or surface mounting (snot).
This is me for what she might be able to do no complicated procedures - as transection of veins and Forwarding of wire.
Sorry for the double post, something stuck.
| |
Nathaniel Tolbert
USA
| |
Posts 9
24 Feb 2011 19:13
| @Thierry - I have an A4000D, A500, A600, CDTV, and an A4000T. I have a copy of Amos Professional but I haven't gotten around to doing anything with it. Research states that it doesn't have true AGA support and there are some iffy patches to allow for that. If I am to learn to program on the Amiga, shouldn't I be learning the language that will give me the most effective usage of resources? I don't think we have met online either. I usually post on Amibay, occasionally on EAB, and I used to post on Amiga.org but don't any more. @Ed Dream - I'm not certain what you mean by purchasing a fee? Could you elaborate more maybe in a specific thread? @Natami Team - Keep up the good work. I am interested to see what the next step in the bring up will be.
| |
Ed Dream
Russia
| |
Posts 28
24 Feb 2011 19:34
| Purchase board.
Why in a separate specific thread?
Offtop?
| |
Thomas Clarke
United Kingdom
| |
Posts 286
25 Feb 2011 10:04
| Nathaniel Tolbert wrote:
|
I take it that there was a change in the I/O pins available with this new FPGA versus the one used in the previous board? Oh by the way first time poster, long time lurker. Aside from the possibility of more I/O Pins and more Logic Elements on the new FPGA, is there any other major differences between this FPGA and the previous one that was used? Also, when trying to look up information on FPGAs I noticed that the speed rating for most isn't posted. Is there a specific reason for this? I wait patiently as I hope to purchase a Natami when it comes out. Sadly I have no developer skills, so all I can do is consume, as I have no other way to contribute to the project.
|
Hi Nathaniel,
Speed ratings can normally be obtained for FPGA devices, but it normally denotes maximum throughput, which isn't useful as a universal measure of final performance. The reason for this is that the performance of FPGA cores depends on their complexity. The more complex the design, the further it gets from the maximum throughput.
For arguments sake, an FPGA core could run at 300MHz. If you had a FPGA core design that was twice as complex, you might find it runs at 150MHz. This is just a simplistic example, but hopefully you get the idea.Two options you have for increasing final 'clock' speed are either using a FPGA which has a higher maximum throughput, and/or you can optimise your FPGA design. This is just my basic understanding of FPGA performance, I'm sure you could get a more complete answer from someone who has programmed FPGAs before, but I hope what I wrote was helpful.
| |
Wojtek P
Poland
| |
Posts 1597
25 Feb 2011 23:08
| Thomas Clarke wrote:
|
Two options you have for increasing final 'clock' speed are either using a FPGA which has a higher maximum throughput, and/or you can optimise your FPGA design.
This is just my basic understanding of FPGA performance, I'm sure you could get a more complete answer from someone who has programmed FPGAs before, but I hope what I wrote was helpful.
|
This is exactly the same as custom chips. No foundry describe their process technology as "2GHz speed", but in terms of gate density, sram density, gate delays etc.
| |
Nathaniel Tolbert
USA
| |
Posts 9
27 Feb 2011 23:17
| @Ed - I wasn't certain what you meant. I will most certainly purchase a board when they are available. I was under the impression that the first boards available would be developer boards and not available to everyone. Is that a wrong assumption?
| |
Ed Dream
Russia
| |
Posts 28
28 Feb 2011 07:18
| Nathaniel Tolbert wrote:
|
@Ed - I wasn't certain what you meant. I will most certainly purchase a board when they are available. I was under the impression that the first boards available would be developer boards and not available to everyone. Is that a wrong assumption?
|
It is common practice.
The first payment will go to the staff team.
The following fees will go to users involved in the development environment - drivers, etc. And I think that as long as the platform will not run out of the box. Ship fee will not be. Therefore, and offers a platform for testing at the user level, and thus want to help the project. Financially of course.
| |
André Jernung
Sweden
| | (MX-Board Owner)
Posts 988
28 Feb 2011 22:56
| *MODERATION POST*
I just removed a couple of posts while stating that this thread, just this one thread, should be free of petty arguments for the sake of people wanting to find information on the Natami bringup process. What happens? The very same individuals as last time start having two pages of new petty arguments in the very same thread. Have you no shame at all?
A new bunch of posts removed. Keep it civil, or don't post at all in this thread.
| |
Amiga Ppc
| |
Posts 246
01 Mar 2011 18:52
| Why can't we all just love each other.... ;o)
| |
Thomas Hirsch
Germany
| | (MX-Board Owner)
Posts 647
03 Mar 2011 23:54
| Currently I am not as fast in the board bringup as I want to be. But there are other also important issues that need to be taken care of. For example the part availability, requests for qupotes and some planning and calculations for an upcoming production run of first boards. Besides the already found layout errors I do not expect further catastrophic errors to be found. In other words, I expect that the current MX layout can be used for a first production run after some minor layout fixes are applied. The very good news is that the memory interface seems to be working. This was my biggest fear. As a first step the CPU driven memory test does not report any error. Of course accurate signal integrity measurements need to be taken. But I do not expect them to show that the MX design is completely wrong. This means that the proof of concept of the hardware side of the 64bit memory interface is successful. The current logic and register design does not make use of the 64bit interface. One of the following tasks is to move all the DMA channels from 32bit to 64bit data width. But this needs to wait till I got the whole LX design running on the MX board and the PCI bus communication is established. PCI is needed for the Flash-ROM, IDE, CF, LAN and USB.
| |
Thierry Atheist
Canada
| |
Posts 1828
04 Mar 2011 00:29
| Let's take this out of context. ;-) :-DDDD
Thomas Hirsch wrote:
|
I expect that the current MX layout can be used for a first production run
|
WOOOOOHOOOOOOOOOOOOOOOOOOOOOOOOOOOOO!!!!!!!!!!!!!!!!!!!
| |
Ed Dream
Russia
| |
Posts 28
04 Mar 2011 07:57
| Important issue the command.
You have not forgot your make and invest in the delivery of the back wall for ITX case?
EXTERNAL LINK
I am referring to this type.
Second question.
Maximum height card with all the features of the module in S-Zorro?
As I understand it, one module installed in natami, will be the top one or more modules to deliver. Sandwich.
For ITX case height is critical. It is necessary to know.
Third question.
Will the boards of passive or active cooling, as the motherboard and on the modules S-Zorro?
And whether they should have?
The fourth question.
Theoretical. What is the estimated maximum power consumption Natami with all modules (S-Zorro) + CF + 1FDD + one 2.5 "HDD, without the PCI card.
Suxx google translate.
Sorry.
| |
Wojtek P
Poland
| |
Posts 1597
04 Mar 2011 13:47
| Ed Dream wrote:
|
First question:
Did you remember to create and invest in the delivery of the back wall for ITX case?
EXTERNAL LINK
I am referring to this type.
Second question:
Maximum height of boards with all the features of the module in S-Zorro?
As I understand it, one module installed in Natami, will be the top one? Or can more modules be stacked on each other, piggybacked.
|
Any amount (to some unknown electrical limit) may be pluggedmost usual config will be one (68060 card) or zero (68050 in FPGA).
| |
Wojtek P
Poland
| |
Posts 1597
04 Mar 2011 13:50
| Thomas Hirsch wrote:
|
The very good news is that the memory interface seems to be working. This was my biggest fear. As a first step the CPU driven
|
Only good engineers fears about potential problems. Bad doesn't ;)It's good to hear you are close to the point you can START it, even with suboptimal performance, and then optimize it circuit by circuit. It only means it will work well :)
| |
Wojtek P
Poland
| |
Posts 1597
04 Mar 2011 13:54
| Ed Dream wrote:
|
Theoretical. What is the estimated maximum power consumption Natami with all modules (S-Zorro) + CF + 1FDD + one 2.5 "HDD, without the PCI card?
|
I don't know, but sure less than 10 watts for everything without FDD and HDD for STANDARD config.
If you will make s-zorro card with intel core i7 it will be more for sure ;)
| |
Bartek "Banter" K.
Poland
| | (Natami Team)
Posts 2277
04 Mar 2011 14:11
| Thank you for the update, Thomas! Very good to learn that 64-bit memory interface is working! Please keep us posted. Good luck!
| |
|
The team will post updates and news here