Uh, I’m pretty sure $135 falls under “exorbitant.” Lol @ “boxed.” How else are you going to ship it? Those aren’t original boxes, for those of you watching @ home.
-ud
This couldn’t have come at a better time as 2 of my friends who live 30 mins away recently got some boards and are in the process of trying to put superguns together. We are wanting to bring these to events that we go to and start up something of our own. The boards they got were from the cabs at X-Mania Europe 3 and they got them signed by Hanashi, Kurahashi, Damdai and NKI, so we can’t wait to run some tourneys and show these off
Just a couple of questions though. When do you expect to start taking orders, and how do I order them (I say them as I may end up ordering the boards for my friends to be delivered to my address)? Also what is the queue looking like right now, and will you treat those who showed interest earlier on in the thread kindly?
I am in the process of compiling a list of (pre)orders (based on date/time of contact). I will contact everyone who has expressed interested in this thread, signed up for the newsletter @ EVO 2012, sent me PMs, sent me e-mails, etc. Those people will then have 2 weeks to respond and confirm an order or forfeit their order position.
As far as SRK related orders, buying and selling in Shoryuken.com is allowed only in the Trading Outlet. After I’ve sorted out everyone who has contacted me thus far, I will open up a sales thread in the Trading Outlet. I cannot say immediately how big the order queue is, as I am still waiting on some responses. Being as I know how many systems I intend to build each month (10), I can at least tell people approximately which month their system should be built in.
Essentially, orders will work like this:
1.) Contact me.
2.) Upon order confirmation (payment not required immediately), I will assign you an order number. Orders will be placed into batches of 10 systems per month.
3.) When I get to building your order, I will send you a payment request. You will then have 2 weeks to send *payment or forfeit your order position (I don’t want to deal with people promising me they will have money in “a few days” and take 5 weeks). Payment can be made in advance, but your order number will dictate when your system will be built.
4.) I will contact you with tracking information when your system has shipped.
*PayPal is acceptable for USA shipments ONLY. If you are outside of the USA, I will accept wire bank transfer, Western Union, etc., Not PayPal. Even if I prove to PayPal that someone’s item has reached their local customs office, the buyer can still claim they never received their item and do a PayPal chargeback, leaving me with nothing. I will not give people that opportunity.
Price is $400, which includes: modified system (CPS2 A Board), power adapter (100-240V AC), video cable (Composite & S-Video), and USB cable for future firmware updates to the controller interface. You will still need a game (CPS2 B Board), TV (CRT with S-Video recommended), and controller (X360 or PS3).
All over the place. Depends on the day, temperature, what shoes you are wearing, etc. General range is $40-$300.
Being as I’m sure some people are going to want to try running their UD-CPS2 with an upscaler, I did some tests on a new Dell monitor I just picked up.
Test Setup:
MC Cthulhu -> UD-CPS2 -> XRGB 2 Plus -> Dell 1908FP
MC Cthulhu is the fastest controller I’ve tested on UD-CPS2. Last time I did latency testing on UD-CPS2, I was getting 1-2ms response with this device. Not that other controllers are crazy slow, but this one is definitely fast.
XRGB 2 Plus is hailed by all to be “lag-free” so, I don’t know what the breakdown is in ms, but I’m going to assume 15ms or less.
I chose the Dell 1908FP for many reasons. It is a large (19") non-widescreen LCD, has decent specs, doesn’t do any funky image stretching (leaves small black bars around the edges), and is readily available. Also of note is that it is speaker bar ready. Dell claims a 5ms gray-to-gray response time for this model. Gray-to-gray is a pretty misleading spec because it could be far from the worst case response time (I believe black-to-white would be worst). The native resolution is 1280 x 1024 and the XRGB 2 Plus sends it 640 x 480, so there is likely some extra processing time for the monitor to double the image (yes, again). In short, I don’t know the exact response time for this monitor, but comparing it to similar monitors, on paper it’s fairly fast.
Video was recorded on a Fujifilm FinePix XP60 @ 60FPS:
NOTE: in both tests, multiple button presses were sampled and in some cases the response time seemed a bit faster than in the examples shown. In this particular test setup, the documented examples are worst case.
This test was not to show an ideal setup, but merely to show how a particular setup performed.
-ud
UD, nice stuff bro!
Just wanted to point it out tho: Keep in mind that in ST, ground normals actually take 1 frame more to startup. If you do the test with an aerial normal (for example) it will probably result in 3 frames of lag, not 4.
Thanks! Yep, JP, Turbo 3. Yeah, I can try those moves, but keep in mind, this is not indicative of actual lag on an arcade cab. The XRGB 2 Plus and the LCD together are adding 1 frame of lag, so far as I’m seeing. If you want real numbers, I could set it up with a CRT.
Thanks! Yeah, as mentioned to Papasi above, this setup is adding a frame of lag anyways, so if an aerial normal is one frame less than a ground normal, I would think on a CRT you would see that test show 2 frames of lag.
-ud
Oh that’s really good news about the MC Chtulhu being so fast. Did you test other PCBs and do you have any results for those? If you get a moment to test and note down the lag of a few different PCBs in order so we can have a recommended setup, I’m sure it would be massively appreciated. I need to get round to installing the Imp and MC Chtulhu in my stick (360 TE). I want the autodetection but I’m not sure which PCB is used as default when the control would be plugged into the UD-CPS2 or a PC? Might have to set the imp up so it’s manual selection via the home button or something.
Also I know the ST Revival guys were doing extensive lag testing with setups in order to find the least laggy emulated version. Not sure what controls they used, but if the Chtulhu is that fast, maybe they should look into it. If they have only been using TE sticks, it might prove even closer to arcade with the Chtulhu
Oh also regarding lag testing how normals come out in-game. Have we ever considered the speed differences of each stage? Like I know some people have done these tests before and say something like “A cr.roundhouse comes out in 6 frames on X setup but 5 frames on Y setup”, but what if one of those setups started on one stage, eg Chun Li’s, and the other setup started on T.Hawk’s or whatever? I wonder if the stages running at different speeds would mean the moves would come out in a different number of frames which would skew the results. I think we need to make sure that normals testing should always be done on the same stage, just to remove another potential variable.
I did some testing of various controllers a long time ago, but I believe my firmware has changed in some timing regards since then, so I’d have to re-test to be certain of the response times with UD-CPS2 today. Inconveniently, in order to do latency testing, you have to actually open up the controller and access the button signals in order to do the test. I don’t mind doing that to my own controllers, but I wouldn’t be comfortable mucking around inside other peoples controllers at an event or something.
Not sure what the mode would be when connecting to UD-CPS2. Would be handy to find out.
Could be but there’s no way to quantify the difference aside from “feel.”
I don’t know enough about the game engine to say if that is of concern or not, but until someone proves it’s not a concern, I think this is a reasonable request. Any requested stage for me to use on future tests?
I don’t know when I’ll get to it, as I really should be working on other things, but since there still seems to be concerns/questions about frames in ST in general, I’ll do some CRT tests.
-ud
Many moves in ST have hidden startup frames that look similar to the character’s neutral pose. I’d suggest using crouch+button or hitting the button while moving so that the start time is clearer when doing lag testing.
Haha, all the more reason to put it off I’m just impressed that with that whole upscaled LCD setup I was running, there was only 1 frame of lag in the Test Menu. I’m wondering if it will go down to 0 frames if I run it straight to a CRT.
-ud
Even with VGA there still scaling and deinterlacing going on, but other image post processing is bypassed.
I just now notice in your photos you are running a XRGB2 Plus. One of the Best way to run RGB (Like in Scart/ Arcade boards 15hz) to VGA (which is RGB HV at 31 hz).
I guess any of the slower stages would work best as long as everyone else agrees to do the same when they test something out so we have reliable results across the board. I’m not sure if this is just a GGPO thing but apparently you can dictate the stage that is selected by the character you pick.
Oh also, I thought of an easy way to check which PCB is enabled via the Imp V2’s auto detect when using it on the UD-CPS2. It’s blindingly obvious so I don’t know why it only just came to me haha. Just have a test setup with a TE PCB and MC Cthulhu wired to the Imp V2 for auto detect, but only connect one button to one PCB and the other button to the other. So if jab is connected to the TE PCB and fierce is connected to the MC Cthulhu, you plug the stick in and whichever move comes out will show you which PCB has been selected when used with the UD-CPS2.
It might not even make that much difference input latency wise, but might as well use the fastest PCB if you have it right?
Thanks! It’s been a long run and now the madness of full production begins! Yay!
What is a “slow” stage?
Sounds simple enough!
In other news, I think I have button config mode such that it should work on any controller now (there were a few @ evo that didn’t work). Also, PS3 button order should be legit on most controllers now (that was mucked up, too, on some controllers). I will go into more detail on this in the future, as I have a plan to improve default button mapping for sticks from various manufacturers.
Going to Saboten Con this weekend, which should allow me lots of testing of the recent PS3 driver changes.
-ud