Overview Features Instructions Performance Forum Downloads Products Reseller Contact

Welcome to the Apollo Forum

This forum is for people interested in the APOLLO CPU.
Please read the forum usage manual.
VISIT APOLLO IRC CHANNEL



All TopicsNewsPerformanceGamesDemosApolloVampireCoffinReleasesLogin
Documentation about the Vampire hardware

Vampire V4 - a Guided Tourpage  1 2 

Ronnie Beck

Posts 33
26 Nov 2019 20:45


Hi All,
 
  I have just put together a short video on the Vampire V4 to whet the appetites of those who pre-ordered or for those who are simply curious to know more about it.  You can find it here:
 
  EXTERNAL LINK 


Willem Drijver

Posts 10
26 Nov 2019 20:52


Nice informative video, thanks!


Vojin Vidanovic

Posts 1421
26 Nov 2019 21:44


Thanks for nice and informative video, added to playlist "Amiga and weird IT" EXTERNAL LINK


Hugo Pereira

Posts 42
26 Nov 2019 22:27


Ronnie Beck wrote:

  Hi All,
   
    I have just put together a short video on the Vampire V4 to whet the appetites of those who pre-ordered or for those who are simply curious to know more about it.  You can find it here:
   
    EXTERNAL LINK   
 

  Excellent short preview! I posted the video on our Vamped Amiga channel on Telegram:
 
  EXTERNAL LINK


Kyle Blake

Posts 40
26 Nov 2019 23:49


Why is the DOOM demo on 1200/030 so slow?

That isn't how it runs. you have a configuration problem.


Sean Sk

Posts 286
27 Nov 2019 00:05


Thanks Ronnie for your awesome review video. Have been waiting a while for a review to come out. Surprised no one else rushed to get one out sooner. Compatibility looks great! I agree that this is the best way to get back into the Amiga if you've been away from it for some time. If I had any doubts about getting one before, those doubts are gone now. Because of limited funds I'll get the V1200 first and then look at one of these later on.

I definitely believe Vampire is the future for Amiga.


Stefano Briccolani

Posts 333
27 Nov 2019 06:22


Cool video!!


Ronnie Beck
(Apollo Team Member)
Posts 105
27 Nov 2019 07:52


Kyle Blake wrote:

  Why is the DOOM demo on 1200/030 so slow? That isn't how it runs. you have a configuration problem.
 

  What do you think is miss-configured?  Could you please elaborate?  An amiga is very simple.  You install your software and run it.  There are no "drivers" for the AGA chipset to configure.  The CPU runs at it's full speed.  Maybe you tried a different Doom port which was optimised better than the version I tried?  It is a bit hard to respond to your speculation here.  Perhaps you could detail your thoughts so that I can check my setup?
 
  I just ran a copy of DOOM that I know works on both the V4 and A1200.  The performance seen is mirrored in the other tests.  I do several side-by-side comparisons.  And what I see makes sense to me.  The V4 has more than 10x the processing power.  The V4 has 60x the chip ram speed.  It makes complete sense that the performance between the two machines is so much.  The V4 is a re-engineered Amiga with many of the bottle necks of an A1200 removed.
 
  The architectual benefits of the V4's design are evident in those demonstrations.  Everything makes sense to me.  But if you know of configuration I should have changed to make Doom run better, please write here.  Otherwise you are just speculating.


Nixus Minimax

Posts 341
27 Nov 2019 07:54


Kyle Blake wrote:

Why is the DOOM demo on 1200/030 so slow?
 
  That isn't how it runs. you have a configuration problem.

Are you sure you are comparing it to the same screen resolution and colour depth? Same executable (there are several Doom ports for Amiga)? Are you comparing to your memory of Doom running on the Amiga or to a recent setup?



Nixus Minimax

Posts 341
27 Nov 2019 07:56


Ronnie Beck, your video is totally awesome! Are you producing videos for a living?



Ronnie Beck
(Apollo Team Member)
Posts 105
27 Nov 2019 08:03


Nixus Minimax wrote:

Kyle Blake wrote:

  Why is the DOOM demo on 1200/030 so slow?
 
  That isn't how it runs. you have a configuration problem.
 

 
  Are you sure you are comparing it to the same screen resolution and colour depth? Same executable (there are several Doom ports for Amiga)? Are you comparing to your memory of Doom running on the Amiga or to a recent setup?
 

I ran the same copy on both machines at the same resolution.  But the V4 achieves a frame rate (~30fps)which is 6 times faster than the A1200 (5fps).  Given that the V4 has at least 10x the CPU grunt, this result matches what one might expect.  If anything, we should expect a bigger performance gap really.

I don't doubt that there is room to optimise the Doom setup.  But doom wasn't really the point of the video.  The point was:  Is the V4 ready for the world?  Is it awesome?  Should I buy one? Yes! Yes! Yes!!!!  :-)


Kyle Blake

Posts 40
27 Nov 2019 09:43


It's not remotely realistic to how people actually use it.
 
  If you play doom on an 030 you don't sit there with some 640x480 mode getting 1fps and watching the screen redraw so slowly it's visible like in the clip.
 
  You run 320x200 with a processor specific c2p routine and get 15-20.

If v4 is so awesome comparisons should be honest.


Nixus Minimax

Posts 341
27 Nov 2019 09:56


The screen resolution used in the video seems to be 320x256. If you can't run this resolution decently on an 030, well, get a V4 and then you can using the exact same executable. That's the point of the comparison, no?

Why compare a stamp-sized screen on the 030 running at 20 fps with an equally stamp-sized screen on the V4 running at 200 fps? That wouldn't make much sense, would it?



Vojin Vidanovic

Posts 1421
27 Nov 2019 10:16


Nixus Minimax wrote:

  Why compare a stamp-sized screen on the 030 running at 20 fps with an equally stamp-sized screen on the V4 running at 200 fps? That wouldn't make much sense, would it?

Well, it proves beefed 080 behaves like 300-500Mhz 030 :)

Surely, 640x or 800x600 should put 080 to sweat, like that v2.7 core running Quake in 640x EXTERNAL LINK


Ronnie Beck
(Apollo Team Member)
Posts 105
27 Nov 2019 12:14


Kyle Blake wrote:

It's not remotely realistic to how people actually use it.
.
.
.
If v4 is so awesome comparisons should be honest.

I ran the same software, in the same resolution on two different machines.  This is fair and it is honest.

But if your sentiment is to defend the honour of the A1200, then I salute your efforts.  But without solid information, your conjecture is just dismissed as speculation.  Thankfully, someone in the YouTube comments has come to your aid and was able to give me the name of a version of Doom that he believed performs this 20fps.  Now I have something real I can look into.


Sascha Wintz

Posts 17
27 Nov 2019 13:20


Great video! Thank you!


Nikos Tomatsidis

Posts 45
27 Nov 2019 14:29


Nice video. Great Production. More like this is nice to promote Vamp.


Dwight Bonney

Posts 9
27 Nov 2019 19:19


Awesome video!!  Can't wait for my pre-order.  I have a semi!!


Kyle Blake

Posts 40
27 Nov 2019 20:06


Ronnie Beck wrote:

    What do you think is miss-configured?  Could you please elaborate?  An amiga is very simple.  You install your software and run it.  There are no "drivers" for the AGA chipset to configure.  The CPU runs at it's full speed.  Maybe you tried a different Doom port which was optimised better than the version I tried?  It is a bit hard to respond to your speculation here.  Perhaps you could detail your thoughts so that I can check my setup?

 
  On the contrary, on an Amiga performance of software that behaves in a way opposed to the hardware design is very much tied to software configuration.
 
  Doom is a software rendered 3D game with chunky pixels. Amiga is of course a planar display intended for 2D graphics with basic hardware acceleration (blitter). Copying the render output to the display therefore requires chunky to planar conversion in software, and the routine you use to do this has to be tailored to your exact specific hardware.
 
  This is absolutely critical. If your C2P routine isn't the correct one for your configuration it becomes a slideshow. If you make this mistake, not even an 060 will save you. C2P is your "drivers" for doom engine 3D.
 
   
 
I just ran a copy of DOOM that I know works on both the V4 and A1200.  The performance seen is mirrored in the other tests.  I do several side-by-side comparisons.  And what I see makes sense to me.  The V4 has more than 10x the processing power.  The V4 has 60x the chip ram speed.  It makes complete sense that the performance between the two machines is so much.  The V4 is a re-engineered Amiga with many of the bottle necks of an A1200 removed.
   
  The architectual benefits of the V4's design are evident in those demonstrations.  Everything makes sense to me.  But if you know of configuration I should have changed to make Doom run better, please write here.  Otherwise you are just speculating.
 

 
 
  You can either run the exact same doom cfg on both A1200 and V4, or you can demonstrate architectural benefits, but you can't do both. Or at least, not well. Running DOOM with wrong C2P routine is like uninstalling GFX drivers on a PC and going "wow! xbox has such good framerate!"

As on Youtube video, DOOMATTACK, C2P routine matching CPU type, small boarder, NTSC 320x200. Doom only runs 200pix tall so extra lines does nothing useful, probably slightly harmful actually.


Gunnar von Boehn
(Apollo Team Member)
Posts 4169
27 Nov 2019 20:18


Kyle Blake wrote:

Doom is a software rendered 3D game with chunky pixels.

 
AFAIK there are several DOOM versions on the market.
I think we can agree on this, right?
 
And those different DOOM version are done by different people and have also different speed.
 
So the very first thing you need to do - is agree on the same DOOM version.
 
Kyle if you say DOOM runs faster for you.
Then this is like saying STAR-WARS has a different story for me than for you.
There is not just Star-Wars movie.
First we need to agree about which episode we talk.

What Ronnie did was fair and correct - he run the same episode of Star-Wars in 2 different cinemas - and compared their view.
This is perfectly correct way of doing it.


posts 23page  1 2