FPGA Based Videogame System

[phoenixdownita]

I'm always conservative with such things, but I can't think of any reason why 500+ couldn't be sold without breaking a sweat as long as it was reasonably easy to purchase and no restrictions were needed on how this was advertised/written about outside of AtariAge.

 

If the goal is tens of thousands of sales or greater, that's a dramatically different level of manufacturing and distribution. If that's the goal, then an existing manufacturing with established connections would likely need to be contacted and partnered with.

 

Kickstarter is always an option (and one suitable for this type of project), but that presents its own issues and challenges. For example, assuming it took off, you'd need detailed manufacturing and distribution plans in place for the best case "worst case" scenario (i.e., being able to cover all that demand). There's a big difference between levels of runs and support that such an operation would need if you're serving 500 -1000 users versus 10,000 - 100,000 users.

My entire point was that to be able to be somewhat profitable it needs some numbers behind it.

In short:

if kevtris were to sell 1000 units at 300US$ and made a profit of 100US$ each, that would net him 100K$, which is barely 2 cores' worth (kevtris mentioned at a point he was seeking 50K for the juicy cores).

[disclaimer: I am not attempting to calculate how much money kevtris could make, that is his business/choice and his alone, I am just putting some numbers to see if it makes sense and what kind of return one could expect ... 30% is likely too high of a ROI]

 

That is why I said what I said. The math needs to add up.

And that is why I am so mad at those RVGS amatuers. If they had a proto and stable specs by now there would be a chance to get 5000K units at a KS, which means targetting 10K+ extra at least after KS, making everyone involved time' plenty worth, and setting up the kind of production/distribution machinery you referred to.

 

The other way would be to make this bare and so cheaper and offload the "mundane" parts [uSB, HDMI out, Networking] to some other board (hence my old suggestion to make it a HAT for RPi2) and advertise it as an accessory there to gain somewhat higher numbers.

 

Krikzz has been making CPLD then FPGA based flash carts for years but I doubt many of them broke the say 20K unit sold [my speculation, I have no data to back this up] because the more interesting are also the more expensive, and the cheaper ones are for system with generally less traction ... it kevtris wants to probe with Krikzz he should get an idea of what kind of retro market really is out there.

Edited September 22, 2015 by phoenixdownita

[Pixelboy]

I'm always conservative with such things, but I can't think of any reason why 500+ couldn't be sold without breaking a sweat as long as it was reasonably easy to purchase and no restrictions were needed on how this was advertised/written about outside of AtariAge.

 

If the goal is tens of thousands of sales or greater, that's a dramatically different level of manufacturing and distribution. If that's the goal, then an existing manufacturing with established connections would likely need to be contacted and partnered with.

 

Kickstarter is always an option (and one suitable for this type of project), but that presents its own issues and challenges. For example, assuming it took off, you'd need detailed manufacturing and distribution plans in place for the best case "worst case" scenario (i.e., being able to cover all that demand). There's a big difference between levels of runs and support that such an operation would need if you're serving 500 -1000 users versus 10,000 - 100,000 users.

There's also the "after Kickstarter" period to consider. Imagine that the system gets great reviews and new potential buyers come out of the woodwork months afterwards and try to order a console, you need to plan ahead for making another run of the hardware. Of course, that's really a "best-case" scenario, but I think kevtris's prototype is good enough that the possibility warrants some consideration.

[+5-11under]

 

So you think it looks better if graphics look like they are made of Lego blocks and could double as a set of stairs (which you'd have to be careful when climbing, lest you gash your shin on the razor sharp edges), than if they look like something far more natural, like a painting on a canvas?

 

 

I like square pixels (or slightly rectangular).

[phoenixdownita]

So you think it looks better if graphics look like they are made of Lego blocks and could double as a set of stairs (which you'd have to be careful when climbing, lest you gash your shin on the razor sharp edges), than if they look like something far more natural, like a painting on a canvas?

 

Yep, that's right. But I understand that others may not feel the same way I do about this.

I don't think many people realize there's quite a few games that are NOT supposed to be viewed at the pixel level. They implicitly used the CRT has a means to generate effects that do not exist at the "pixel" level.

 

Case in point Tower Topple for atari 7800.

Please take a look at the SVideo version and at the composite version, the SVideo version (or emulated version if you prefer) is simply WRONG in the way it looks,

http://videogamecritic.com/7800sz.htm?e=25333#rev608

the tower and the eye are not supposed to look like vertical stripes/bars and on composite (and even better on CRT) in fact they blend to generate solid colors.

It was called "artifacting" and you can read a lot about it here:

https://en.wikipedia.org/wiki/Composite_artifact_colors

 

Another example:

http://videogamecritic.com/xemp.htm?e=25333#rev4262

[don't think for a moment the game was meant to be seen each other pixel and with those weird colors]

you can ask here on the Atari 8 bits forum, there's a wealth of info about how the various Atari 800 (XL) differed in the way the generated artifacts it's not even funny.

 

But I agree that blurred images are not nice to look at that but also laser cut ones, they both look unnatural.

Edited September 22, 2015 by phoenixdownita

[Keatah]

I didn't get the "BLIP" option. I must be stupid at the moment.

[MaximRecoil]

The blocky artwork was due to a technical limitation. Without the limitation, the graphics would have most likely looked like hand-drawn comic book art, just like the accompanying artwork for many games (on labels, boxes, manuals, sideart, etc). Fortunately, 15 kHz CRT displays are inherently the world's greatest anti-aliasing devices, simply as a byproduct of design, which mitigated the eyesore of the Lego graphics. Since it is generally agreed that blatantly pixelated images are ugly, anti-aliasing is extensively used, even more today with all of the digital displays in use than 10 or 15 years ago when CRT PC monitors were more common. Whenever you resize an image in a graphics editor, by default it uses a filter with an anti-aliasing algorithm. All modern web browsers anti-alias text by default (they never used to, because CRT PC monitors don't need it for small text), and Windows has anti-aliased its UI text by default since XP.

 

High resolution displays, whether they be in the form of CRT or digital, effectively "turns off" the inherent anti-aliasing that 15 kHz CRTs have, revealing the ugly pixelization. Images are all about illusions. The closer you examine any image, the less it looks like it is supposed to look. For example, look closely at a picture in a magazine, and this is what's actually there (rosette halftone pattern from the offset printing press, consisting of 4 colors [CMYK]):

 

 

But when viewed normally, as it was intended to be viewed, it looks like water:

 

 

Likewise, old games were intended to be viewed on a ~15 kHz CRT display.

Edited September 22, 2015 by MaximRecoil

[telengard]

I didn't get the "BLIP" option. I must be stupid at the moment.

I was thinking of those old Tomy games, but those are electomechanical IIRC.

[Keatah]

Ahh yes I thought maybe t'was that. Well, it would be an interesting project to do a 3D rendered simulation of the mechanics inside.

[MaximRecoil]

 

I like square pixels (or slightly rectangular).

 

I like not seeing distinct pixels at all, given that they were never intended to be seen, and have no relationship to any artform man ever came up with prior to it being forced by technical limitations of low-powered computer hardware. Since CRTs have no pixels, but rather, round phosphor dots, the problem of harsh/blatant pixelization doesn't exist on them, at least not on ~15 kHz CRTs. The problem does exist on high-resolution CRTs (e.g., CRT PC monitors) with a tiny dot pitch, because their phosphor dots are so small that they can create the illusion of razor sharp corners at a normal viewing distance, making them not much better for displaying vintage graphics than modern digital displays.

Edited September 22, 2015 by MaximRecoil

[telengard]

Oh, forgot to mention the price for me isn't a big factor. On going support (via the community, whatever) is most important, also the ability to map inputs for the systems that way I could use my USB retro controller adapters would be excellent! SD card is perfect and having composite would be nice so I can connect it to my 1702.

[+5-11under]

 

I like not seeing distinct pixels at all, given that they were never intended to be seen, and have no relationship to any artform man ever came up with prior to it being forced by technical limitations of low-powered computer hardware. Since CRTs have no pixels, but rather, round phosphor dots, the problem of harsh/blatant pixelization doesn't exist on them, at least not on ~15 kHz CRTs. The problem does exist on high-resolution CRTs (e.g., CRT PC monitors) with a tiny dot pitch, because their phosphor dots are so small that they can create the illusion of razor sharp corners at a normal viewing distance, making them not much better for displaying vintage graphics than modern digital displays.

 

Just my opinion...

If I'm running HDMI on a 32" or 42" TV, the amount of blur would be brutal. I design pixels with real (or virtual) graph paper to be square, not rounded or blurred. In my mind, pixels are supposed to be square, and any TV that doesn't display them as square is inferior (yes, I understand there could be issues/compromises, but that works both ways). I like pixels, and if I can see them properly, so much the better for my gameplay - I hate the fuzziness of new games where I don't know if my player is going to be "hit" or not.

/don't shoot me.

[Asbrandt]

The niche within a niche effect is a real problem indeed, but I think Kevtris is really looking for an idea of the best balance between price and features here before posting in more places for a much wider interest check to determine if it's a viable niche to fill.

 

I'm hoping for a favorable result ofcourse, having wanted this thing since the first time I read about it quite some time ago, but I may be optimistically biased since I am part of the exact target audience.

Edited September 22, 2015 by Asbrandt

[phoenixdownita]

Just my opinion...

If I'm running HDMI on a 32" or 42" TV, the amount of blur would be brutal. I design pixels with real (or virtual) graph paper to be square, not rounded or blurred. In my mind, pixels are supposed to be square, and any TV that doesn't display them as square is inferior (yes, I understand there could be issues/compromises, but that works both ways). I like pixels, and if I can see them properly, so much the better for my gameplay - I hate the fuzziness of new games where I don't know if my player is going to be "hit" or not.

/don't shoot me.

Not all the time it was the case:

http://www.chrismcovell.com/gotRGB/screenshots.html

check the Genesis section and look how via clever tricks (the waterfall in Sonic is a glaring example) they achieved effects that are NOT possible at the pixel level on that hardware (the emu shots look wrong because the gfx artists wanted the actual composite signal + CRT to do the blending part as they had no other way to achieve it).

 

I can understand that in modern day on modern hardware such tricks are no longer needed (enough colors to not need dithering or artifacting, and real transparency to avoid "each other pixel" tricks) but old games sometimes relied on that so the statement about "wanting to see the pixels" only applies to what you produce not in general.

Edited September 22, 2015 by phoenixdownita

[Radfoo]

Just thinking about my vote, think I said 8bit and 16bit although 8bit is probably more my thing. Would be very happy to purchase as it is with the cores you already have rather than wait. Been following yours and others work with FPGA seemingly for years, missed MiniMig, tried to buy an FPGAArcade a couple of years ago and currently considering a MIST board, though yours isn't released it seems more complete.

Edited September 22, 2015 by Radfoo

[MaximRecoil]

If I'm running HDMI on a 32" or 42" TV, the amount of blur would be brutal.

 

What blur? HDMI is only found on high-resolution displays, and the problem with them is not blur, it is that primitive graphics look like Lego blocks.

 

 

 

I design pixels with real (or virtual) graph paper to be square, not rounded or blurred.

 

You don't design them to be square, they are inherently square. And no one wants a blurry image anyway.

 

 

 

In my mind, pixels are supposed to be square, and any TV that doesn't display them as square is inferior

 

Pixels are square, but the problem with that is, large squares are a poor tool for making aesthetic drawings. This is why people have always drawn with e.g., a pen or pencil rather than a square rubber stamp. So in this case, a display that inherently rounds the edges of the blocks results in a more aesthetic presentation of the drawings which were hamstrung by limited technology in the first place. It is the exact same reason that anti-aliasing is used so much these days. Which of the following two renderings of the letter "R" do you prefer?

 

 

Edited September 22, 2015 by MaximRecoil

[BigO]

 

I like not seeing distinct pixels at all, given that they were never intended to be seen, and have no relationship to any artform man ever came up with prior to it being forced by technical limitations of low-powered computer hardware.

Maybe pointillism, but I don't think it's generally done with little squares in orderly formations.

[+5-11under]

 

What blur? HDMI is only found on high-resolution displays, and the problem with them is not blur, it is that primitive graphics look like Lego blocks.

 

You don't design them to be square, they are inherently square. And no one wants a blurry image anyway.

 

Pixels are square, but the problem with that is, large squares are a poor tool for making aesthetic drawings. This is why people have always drawn with e.g., a pen or pencil rather than a square rubber stamp. So in this case, a display that inherently rounds the edges of the blocks results in a more aesthetic presentation of the drawings which were hamstrung by limited technology in the first place. It is the exact same reason that anti-aliasing is used so much these days. Which of the following two renderings of the letter "R" do you prefer

 

What blur? The blur I would get if I tried to get my 42" TV to show the dithering or other CRT effects. I guess that wasn't clear. I prefer the so-called "Lego block" look.

 

Square pixels on graph paper: it sounds like we're on the same page here.

 

The letter R: There's nothing wrong with high resolution, but if a console's resolution is low (what's that "R", about 64 pixels vertically, or 1/3 of the vertical resolution of a ColecoVision?), then I want to see the first one - crisp and square pixels. I don't want rounding, blending, or any other effects or processing. I'm not comparing Mondrian vs. Monet vs. Michaelangelo - I love them all. I just want square pixels when my classic low resolution video games are scaled up onto high resolution displays.

 

/that's just me.

[Pixelboy]

/that's just me.

And me.

[+Tarzilla]

 

What blur? The blur I would get if I tried to get my 42" TV to show the dithering or other CRT effects. I guess that wasn't clear. I prefer the so-called "Lego block" look.

 

Square pixels on graph paper: it sounds like we're on the same page here.

 

The letter R: There's nothing wrong with high resolution, but if a console's resolution is low (what's that "R", about 64 pixels vertically, or 1/3 of the vertical resolution of a ColecoVision?), then I want to see the first one - crisp and square pixels. I don't want rounding, blending, or any other effects or processing. I'm not comparing Mondrian vs. Monet vs. Michaelangelo - I love them all. I just want square pixels when my classic low resolution video games are scaled up onto high resolution displays.

 

/that's just me.

I'd prefer a config file on the SD as part of the core that lets me turn on the TV Scanline/other filters on or off depending on system.

[Asbrandt]

If the effect that produces 'artifact colors' is predictable and consistent enough to be represented mathematically regardless of source system, it should be possible to simulate to a satisfactory level with minimal latency for the systems that call for it (alot but not all NES and Genesis, probably some home computers and alot of pre-NES but I have no experience with either of those) since it only occurs on the horizontal axis.

 

I would guess that plus scanlines would be the limit of "tube simulation" options, although that would definitely be more than enough for me.

 

HQX is something else entirely and to be honest I am not the biggest fan, even if it is interesting from a technical perspective that it was implimented on the HiDefNES.

[leods]

Not sure what to say here. I would already have bought a retron 5 if not for the sketchy stuff they did, their very dishonest Advertising and poor build Quality.

 

I have a Mega Drive, and it's not even connected right now, because I lack space. If I could have a System to Play old cartridges from different Systems on my PC Monitor, that only required power supply and an HDMI cable I'd definitely buy it for 200 bucks the base unit plus a bit more for SNES, MD, NES/Famicom and GB.

 

USB Controller is a GREAT idea. There are many Adapters for the old Controllers, and there are more than a couple amazing USB Controllers out there, like the Hori Gem Pad and Arcade sticks from different manufacturers. it's also pretty easy to mod something to Play on USB.

 

Wouldn't it be possible for People to get this board, plan a custom core and develop some new games, maybe commertially, maybe just for fun? How could the RVGS Team screu up so bad? They had everything ready. If they just let go of the damned Jag cases and helped you finish your Project this Thing would already be kickstarted. Damn man.

 

If you Need, maybe get a Team with you to help, but considering you've already made a ton of stuff by yourself, I'd say you should start preparing to make this Commercial. I don't think you'll sell a ton, but since you're doing the for for free, just for fun anyways, might as well just make your buck and let us have some fun.

[Bar81]

SFC/SNES, MD/Genesis, PCE/TG-16 and Neo Geo would be incredible. If those are all in, I'm in for a couple of units assuming it's not breaking $300 a unit.

[Asbrandt]

What I find interesting about the responses is that theres a wide variance not just in desired price/features but in the price:feature ratio.

 

That might make things tougher, since there seems to be a fair amount looking for less than the likely target price mentioned in the OP, perhaps influenced by the Retron 5's price even though this serves a much more specific niche.

[leods]

I'll be sincere, I think there is a flaw in the questionary, because there is a question about the Price Point, but it's not relating that to the Features People expect.

 

I marked I would pay 200 bucks. But if this Thing actually does SFC/SNES, MD/Genesis, PCE/TG-16 and Neo Geo, plus all the stuff it already does, I'd pay 300 bucks easy. Damn, if it's plug and play, this Thing would sell like water in the desert amongst Retro Gamers.

 

But yeah, certainly the 150 bucks pricepoint of the retron 5 makes things harder for the Project, something Kev certainly took into consideration from the start.

 

This has way more potential though..

[Kaijuseijin]

 

 

I marked I would pay 200 bucks. But if this Thing actually does SFC/SNES, MD/Genesis, PCE/TG-16 and Neo Geo, plus all the stuff it already does, I'd pay 300 bucks easy. Damn, if it's plug and play, this Thing would sell like water in the desert amongst Retro Gamers.

 

 

These are my feelings as well. Different price points for the amount of systems it can reproduce. Something along those lines.