Excitement about NatAmi

[commodorejohn]

Right, right. Guess we'll see how it shakes out, hopefully soon!

[the_leander]

You continue to troll.

And you continue to fail to provide evidence to backup your assertions.

You failed algebra.  The potential is closer to 400x the speed of an A500 blitter but for other reasons that will only end up in the 100x-200x range.  100x is a conservative #.

Citation needed.

The point you keep failing to take on board is that doing a straight calculation only works if you're talking about the two blitters being functionally identical and no different in terms of how they do their job.

The moment you start throwing in buffers, 32bit blitting etc. All bets are off.

This is of course assuming that the blitter on the natami actually works as advertised and doesn't have some hitherto unknown bug etc.

That is all.

But again, this is not fact. Without benchmarks on real (or as real as an fpga powered platform can be) hardware, you cannot prove your case.

As another in the thread said - what you're doing is speculating. Only you're claiming it as "fact", which is dishonest.

You could challenge them if you had facts but you don't even know the first thing about blitting and memory access speeds.

I can challenge anyone's claims at any time as can anyone else on this public forum. Those who actively refuse to provide evidence and are, time and again spanked for talking rubbish.

Ad hom

Ah, run out of arguments so you attack the person asking you for proof.

What was it you said? Ah yes

If you want to call me a liar,

No I accused you of being intellectually dishonest. Which you have been. Reading comprehension fail.

then come back with facts instead of challenges.

Again, it is not my place to disprove your claims.

 I gave you facts

You gave nothing. You pointed to a wikileaks article and then said "do the math" without any evidence to show that the blitter on natami even works.

See my explanation above as to why what you said was and continues to be a complete and utter nonsense.

Rogue on AW.net said they OS4 could run on the Wii.  You, sir, lack much upstairs...

Coming from you, that means absolutely nothing, mr gamecube. You were running around the forums talking about porting to the gamecube first, later the wii, without even the slightest thought as to how wise a choice that would be. As JJ said, rouge said it to get you to shut up.

The only thing that was going through your head, was that it had a PPC. Same as when you tried to imply that the kurobox nas might be a good fit for OS4, even with it lacking a graphics chip or the means to add one.

[Hattig]

Citation needed muppet.

Your posts in this thread have been abrasive already, but now you're just being rude and obnoxious.

I hope a mod bans you for a while to teach you a lesson.

[the_leander]

Your posts in this thread have been abrasive already, but now you're just being rude and obnoxious.

Very interesting that you ignore his multiple ad homs previous to this and focus in on that one. Very interesting indeed.

I hope a mod bans you for a while to teach you a lesson.

If and when I breach the TOS, I'm sure I'll receive the appropriate infraction.

[Hattig]

The point you keep failing to take on board is that doing a straight calculation only works if you're talking about the two blitters being functionally identical and no different in terms of how they do their job.

His point is very clear, and is based upon simple assumptions:

1) The NatAmi blitter is backwards compatible - it's not much of an Amiga hardware re-implementation if it isn't

2) The NatAmi blitter works (I mean, seriously, you have issues regarding this when the hardware is FPGA and can easily be updated if there are latent bugs?)

The maximum amount of data that can be blitted is limited by the memory bandwidth. It is further limited by how fast the blitter runs, but I think we can safely assume that it will be running faster than on the original hardware.

The moment you start throwing in buffers, 32bit blitting etc. All bets are off.

Buffers (for DDR2 memory bursts) - speed up the blitter
32-bit blitting - speed up the blitter

These are both arguments in favour of Lou's assertions, and I presume these are also factored in to the NatAmi team's projections. It's not like 100x faster blitter is some unattainable magical thing - the AGA blitter was appalling anyway, apparently a 4x speedup would have been trivial if it had supported 32-bit and page mode on the memory! That's before you try running it faster than ~4MHz...

This is of course assuming that the blitter on the natami actually works as advertised and doesn't have some hitherto unknown bug etc.

This is not a valid argument. If the blitter doesn't work, no-one will buy a NatAmi. It's not like they didn't show early prototypes running on the C-One and the other evaluation board is it? And being an FPGA, rare occurrence bugs can be fixed when they're noticed.

(rudeness removed) Without benchmarks on real (or as real as an fpga powered platform can be) hardware, you cannot prove your case.

Ah, the FPGA isn't hardware argument. Sheesh.

As another in the thread said - what you're doing is speculating. Only you're claiming it as "fact", which is dishonest.

At least he has made an effort to back up his claims.

I can challenge anyone's claims at any time as can anyone else on this public forum. Those who actively refuse to provide evidence and are, time and again spanked for talking rubbish.

You can argue like an adult, or just call people names. You call people names. Grow up.

Ah, run out of arguments so you attack the person asking you for proof.

You appear to have projection issues.

*more personal attacks removed*

Coming from you, that means absolutely nothing, mr gamecube. You were running around the forums talking about porting to the gamecube first, later the wii, without even the slightest thought as to how wise a choice that would be. As JJ said, rouge said it to get you to shut up.

I think I get it. You don't care what Lou is saying, you just want Lou to shut up because he is Lou. That's very personal. Why are you dredging up something that apparently is YEARS OLD? I guess you have a personal vendetta. Take it away from this forum please.

[Karlos]

Gentlemen, a spot of calm if you please.

Until I have a working NatAmi in my grubby paws, as far as I am concerned it's all speculative. On that point, a modern implementation of the hardware, if it lives up to the promise of being able to utilise the full DDR2 memory bandwidth would seem to qualify as much, much faster than the original. However, that's as much as can be said. Putting numbers on it at the moment, no matter how educated a guess, remains a  guess.

[jj]

I think I get it. You don't care what Lou is saying, you just want Lou to shut up because he is Lou. That's very personal. Why are you dredging up something that apparently is YEARS OLD? I guess you have a personal vendetta. Take it away from this forum please.

 
Who made you a mod ?

[jj]

Gentlemen, a spot of calm if you please.
 
Until I have a working NatAmi in my grubby paws, as far as I am concerned it's all speculative. On that point, a modern implementation of the hardware, if it lives up to the promise of being able to utilise the full DDR2 memory bandwidth would seem to qualify as much, much faster than the original. However, that's as much as can be said. Putting numbers on it at the moment, no matter how educated a guess, remains a guess.

 
Karlos be careful, how dare you say its just a guess, its just moving memory.   Did you fail maths in school etc etc

[the_leander]

His point is very clear, and is based upon simple assumptions:

Which he claims as fact. Do try to keep up.

My problem isn't with the assumption as such (though given the proposed updates in the rest of the enhanced AGA being proposed by the Natami team I think a simple maths equation just won't cut it), but with the fact that he states it as fact and then gets angry when asked to provide evidence.

I'm not arguing that having two essentially identical bits of hardware, with one working many times faster than the other will not be faster, I'm saying that until we have working hardware in our hands, we cannot say accurately "how much".

Ah, the FPGA isn't hardware argument. Sheesh.

That wasn't what I was getting at. Truthfully I have no pony in that argument.

At least he has made an effort to back up his claims.

I'm not making any claims though. I'm asking where the evidence for this thing actually doing what is being said is.  

If it had been prefaced with "in theory" or "is projected to have" or some other similar disclaimer, I would have said very little on the matter.

You can argue like an adult, or just call people names. You call people names. Grow up.

All I did was ask for evidence "Citation needed". Apparently these days that's abrasive, rude and obnoxious. Fair enough.

I think I get it. You don't care what Lou is saying

Still failing. Read above.

My reason for "dredging up" the past was to point out that far from some pedestal occupying god, to be gazed at in wonder but not to be questioned that he'd apparently like to be treated as. He is in fact only a hair's breadth from being Atheist2. Honestly I don't know or really care about him, in fact the word "muppet" pretty much sums up my feelings toward him in their totality.

[Karlos]

Karlos be careful, how dare you say its just a guess, its just moving memory.   Did you fail maths in school etc etc

If there's one thing I've learned when doing low level work with real hardware, there's the theoretical speed of which the hardware is capable and the actual speed that is attained in the real world with applications that test it. The interface between software and hardware is often a murky place.

User
|
Application
|
OS Graphics API
|
Driver
|
Hardware

Remember, to use the blitter in an OS friendly manner, you have to wait for it, own it, set it up to do your operation, disown it etc. etc. All of those steps take time, which is more or less independent of how fast the blitter itself can do the operation you've asked.

Will you see a 100x speed up in a real world application? It all depends. If the setup cost is small, maybe you'll experience more than that. OTOH, if there are many small blits going on frequently, then the setup latency may dominate and you'll observe a lot less.

[Hattig]

Which he claims as fact. Do try to keep up.

You have an attitude problem. You showed it in your early responses to lou's post - e.g., "I'd owe you sweet FA chum." - a very aggressive phrasing in English, it's effectively one step before someone glasses someone in the face in a pub.

When someone writes "100x faster blitter" they're not being 100% pedantically accurate. In English it it understood as "around 100x faster" to "a boat load faster". He then went on to mention that FPGAArcade is getting 40x faster blits (although he could have cited that, but again, the forums on Amiga.org aren't wikipedia, they're chatty forums), that NatAmi's specs are even higher, so it isn't an unreasonable expectation.

There's nothing wrong with Lou stating what the NatAmi team have stated as their aims. You should be asking for evidence from SamuraiCrow or other NatAmi people as to the speed and status of NatAmi - however to their credit they have said that they'd rather work on NatAmi right now than give continuous updates.

It is interesting to note that you jumped on Lou's 100MHz 68060 comment very quickly, and that in the next comment SamuraiCrow said that it was actually a 99MHz 68060, but that the 68050 core would be over 100MHz.

My problem isn't with the assumption as such (though given the proposed updates in the rest of the enhanced AGA being proposed by the Natami team I think a simple maths equation just won't cut it), but with the fact that he states it as fact and then gets angry when asked to provide evidence.

He tried to show you why he thought that, basically faster FPGA than FPGAArcade, a faster memory bus, etc.

All I did was ask for evidence "Citation needed". Apparently these days that's abrasive, rude and obnoxious. Fair enough.
Still failing. Read above.

No, that's not what is abrasive, rude and obnoxious. It's what and how you are writing it that that is.

My reason for "dredging up" the past was to point out that far from some pedestal occupying god, to be gazed at in wonder but not to be questioned that he'd apparently like to be treated as. He is in fact only a hair's breadth from being Atheist2. Honestly I don't know or really care about him, in fact the word "muppet" pretty much sums up my feelings toward him in their totality.

So he had some fancy ideas in the past that might have been based entirely upon the Gamecube/Wii having a PowerPC processor.

Instead of responding to him if you hold him in low regard, why not just ignore him? Or accept that some people might have less technical knowledge, and that their desire is to have a friendly discussion in an online forum in a similar way to having a friendly chat down the pub. It's meant to be fun. He's not advocating genocide for MorphOS users or something! It's been pointed out loads of time in this thread that we'll find out in due course what NatAmi is like, we're all working from what has been written on the NatAmi website, said in the NatAmi forums and said by NatAmi members here and elsewhere.

As for me, a 100x faster blitter sounds great, but I won't cry if it's only 50x faster in the end due to the FPGA, the design, the memory speed or the weather in Iceland on that day. The same goes for the 68050 design they're working on. I wish them the best of luck but we'll see if it will clock at 133MHz, or 66MHz. If they know it will, they could throw us a tidbit of information now, but until then we're taking their word for it in order to have a fun chat about this project.

[Hattig]

Who made you a mod ?

No one, I'm just saying that I don't want to read continuous personal attacks in an otherwise fun thread about NatAmi and what it could mean, in a forum that has the word 'ideas' in the name.

[Louis Dias]

It is interesting to note that you jumped on Lou's 100MHz 68060 comment very quickly, and that in the next comment SamuraiCrow said that it was actually a 99MHz 68060, but that the 68050 core would be over 100MHz.

He chooses to ignore facts.  The LX board was shown in video to load and decode 24bit 256kish images in HAM8 in real time in around 5 seconds.

He tried to show you why he thought that, basically faster FPGA than FPGAArcade, a faster memory bus, etc.

Again he chooses to ignore facts while providing no valid counter arguements other than it doesn't exist...despite the actual existence of the LX board.

So he had some fancy ideas in the past that might have been based entirely upon the Gamecube/Wii having a PowerPC processor.

It was simple logic.  GC/Wii use PPC and have well know hardware (see Wiibrew.org).  A1 was dead and buried with no new hardware in sight.  Though not ideal, it puts an OS4 box in everyone's living room.

Instead of responding to him if you hold him in low regard, why not just ignore him? Or accept that some people might have less technical knowledge, and that their desire is to have a friendly discussion in an online forum in a similar way to having a friendly chat down the pub. It's meant to be fun. He's not advocating genocide for MorphOS users or something! It's been pointed out loads of time in this thread that we'll find out in due course what NatAmi is like, we're all working from what has been written on the NatAmi website, said in the NatAmi forums and said by NatAmi members here and elsewhere.

As for me, a 100x faster blitter sounds great, but I won't cry if it's only 50x faster in the end due to the FPGA, the design, the memory speed or the weather in Iceland on that day. The same goes for the 68050 design they're working on. I wish them the best of luck but we'll see if it will clock at 133MHz, or 66MHz. If they know it will, they could throw us a tidbit of information now, but until then we're taking their word for it in order to have a fun chat about this project.

Luckily I'm not the type to file reports.  His ignorance is here for all to see.

[the_leander]

You have an attitude problem.

So does everyone you don't like.

You showed it in your early responses to lou's post - e.g., "I'd owe you sweet FA chum." -

And if you look carefully, you'd see that there was a link explaining the meaning behind the "You are X and I claim my £5" from wikipedia.

Context, will you bother to supply any?

a very aggressive phrasing in English, it's effectively one step before someone glasses someone in the face in a pub.

Maybe it is considered "aggressive phrasing" amongst the chattering classes, but down here on the front lines it's every day language used as a polite replacement for "Fuck All", the verbal equivalent of putting stars * in place of letters. In short, you're talking nonsense.

When someone writes "100x faster blitter" they're not being 100% pedantically accurate.

And instead of saying anything like you're implying, he went off on one about it being pure maths to prove his case that what he said was in fact, a fact. One which if I didn't agree to, would mean I was subnormal.

You're determined to ignore everything that was said in its context to prove your point, aren't you?

It is interesting to note that you jumped on Lou's 100MHz 68060 comment very quickly, and that in the next comment SamuraiCrow said that it was actually a 99MHz 68060, but that the 68050 core would be over 100MHz.

SamuraiCrow had the good grace (and I mentioned it earlier in the thread too) to back himself up with evidence and preface any statements carefully with the point that the hardware wasn't ready yet.

You'll note my response to him was very different.

So he had some fancy ideas in the past that might have been based entirely upon the Gamecube/Wii having a PowerPC processor.

LMAO. He was spamming thread after thread both here and on AW.net for months about it being the ideal thing despite having had it explained to him in calm, patient manner why it wasn't such a hot idea.

After that, I'm not going to really take him all that seriously when it comes to hardware.

As for me, a 100x faster blitter sounds great, but I won't cry if it's only 50x faster in the end due to the FPGA,

It's all good.

[jj]

Anyway lets remember that winning an argument on the internet and special olympics etc

[Louis Dias]

If there's one thing I've learned when doing low level work with real hardware, there's the theoretical speed of which the hardware is capable and the actual speed that is attained in the real world with applications that test it. The interface between software and hardware is often a murky place.

User
|
Application
|
OS Graphics API
|
Driver
|
Hardware

Remember, to use the blitter in an OS friendly manner, you have to wait for it, own it, set it up to do your operation, disown it etc. etc. All of those steps take time, which is more or less independent of how fast the blitter itself can do the operation you've asked.

Will you see a 100x speed up in a real world application? It all depends. If the setup cost is small, maybe you'll experience more than that. OTOH, if there are many small blits going on frequently, then the setup latency may dominate and you'll observe a lot less.

Theorectical maximums are the only thing worth comparing because all the inefficiencies you mention also apply to the old hardware.  Just by adding buffers, yarube has made the TG68 core run like an '040...yet I'm not allowed to cite that as co-oberating evidence?  Natami bus is about 4x faster than MikeJ's board directly due to DDR2 vs. DDR1.  It's fpga is also superior.  These are facts.

One of the factors that helps make full '020's, '030's etc... faster is cache.  The '050 core is getting, iirc, 32k of data and instruction cache.  This is also what yarube did to the TG68 core.  It's not rocket science to people who know what they are doing.  Gunnar and Thomas have real world industry experience.  I see too many people here belittling their work.  It annoys me, especially when some of those people are clearly threatened by the product because they are a developer for an OS who's market could be undermined by it's success.  Other people, **cough**the_leander**cough**, just aren't too bright.