"MS4" wishlist thread

[racingmini_mtl]

You also have to consider that you're buying the rights to the firmware, development of said firmware, and support from the developers and vendors. Because you can buy your own board and components but you'd still need to spend a lot of time and effort on the rest before you have got an ECU and not just a pile of electronics.

By the same token, DIY sells a MS3/X + RTC + knock module for $810, and it's a ready-to-roll unit. The MS3Pro Module, at $850, still has to be fitted with a carrier board and mounting hardware, an enclosure (which may or may not require significant machining to accept the boards and include the openings), and connectors. For a full-on drag car with a V8 plus nitrous and a turbo, it may well be the non plus ultra. For someone with a simpler installation, it's over-horsed, over-designed, over-sized, and over-priced.

Don, that's what I was commenting on:

I feel that is way overpriced for what it is, a PCB and components.

I've already mentioned my take on the rest.

Jean

[elaw]

By the same token, DIY sells a MS3/X + RTC + knock module for $810, and it's a ready-to-roll unit. The MS3Pro Module, at $850, still has to be fitted with a carrier board and mounting hardware, an enclosure (which may or may not require significant machining to accept the boards and include the openings), and connectors. For a full-on drag car with a V8 plus nitrous and a turbo, it may well be the non plus ultra. For someone with a simpler installation, it's over-horsed, over-designed, over-sized, and over-priced.

Well the "classic" MS3 isn't *quite* ready-to-roll, you do have to install the RTC and knock modules yourself.

But I have to say other than that I agree with the above argument. Once you've put in the minimal effort to install those two add-on cards, you have a complete "product" that's ready to install and use. With the pro module you have to add a substantial amount of time & materials in order to get to the same place.

And the kicker is the pro module should be considerably easier (=cheaper) to manufacture. It's one PCB instead of five, there are no inter-PCB interconnects, few through-hole parts, and no case. And while I have tremendous respect for the effort that goes into the development and support of the firmware, that's not really germane to this comparison because they both run the same firmware!

Put it this way... if you went to your local wireless place and saw they were selling a "module" version of your cellphone with no display and no case, would you expect to pay more for it?

But I will also say that I think the pricing of the MS product line in general is very favorable compared to other comparable products. It's just that the pro and pro module aren't *as* favorable as some of the other things.

Maybe it's like my local sandwich shop that charges more for a grilled cheese sandwich than for a cheeseburger. I'm not quite sure how deleting the beef patty from the sandwich makes it more expensive, but some things just is they way they is.

[Rickster]

Volume and market share is a powerful thing, but at the price point MS3-pro (or the module) is at, it won't happen.

10 units @ $400 profit < 100 units @ $50 profit

2 more cents.

Rick

[jsmcortina]

Volume and market share is a powerful thing, but at the price point MS3-pro (or the module) is at, it won't happen.

Too cheap compared to the competitors? I was going to name some here, but I won't. Please do look up competitors ECUs with similar features and see how they compare.

James

[arran]

The reason I bought an MS3 is because of the value for money, I would respectfully disagree with those in this thread saying it is over priced. Saying that however, as a consumer I would of course like more for less!
Also the need for low z injector drivers I think is going to lessen as time passes because of the recent availability of EV14 style high impedance injectors, they are cheap and come in substantial capacities.
Initially one of the reservations I had with MS3 was the DB connector but after using the computer for a couple of years I don't have a problem. They are reliable, cheap, simple.

What I would like my MS3 to do that it does not:
1) The feature I would like added most is more ADC. The existing CPU has quite a few more that are not presented.
2) Smaller physical size (I'm OK with the current MS3 tall case in my install but smaller would be better)
3) Drive by wire to allow modern traction control and more elegant idle control
4) The world wants internet connectivity, inbuilt Wi-Fi would be cool
5) How about live telemetry over mobile phone?

Asking myself the question "what has been the hardest thing about setting up my MS3?"
As a user I want to get it working for as little effort as possible, I do NOT mean this as a criticism to the brilliance of the hardware/firmware developers but I have had a great deal of difficulty with something that turned out to be a bug. I would therefore prefer that more time be put into validating that the current stable firmware is in fact stable. I am aware that effort is being spent on at least two major firmware variants, 1.4/1.5, with new features which is much more interesting as a developer than bug checking the existing version.

[jsmcortina]

What I would like my MS3 to do that it does not:
1) The feature I would like added most is more ADC. The existing CPU has quite a few more that are not presented.

Take a look at the Microsquirt IO-box.

I would therefore prefer that more time be put into validating that the current stable firmware is in fact stable.

I hear you, but it seems very difficult to get in-depth tester feedback, much as we'd like it.

James

[jsmcortina]

While mentioning prices I just did the sums - using retail prices:
MS3-Pro inc 8' harnesses ($1199) + MAP sensor ($64) + pigtail ($11.50) = $1274.50

MS3XV357 ($659) + harnesses ($79 + $79) + knock ($85) + RTC ($65) + assy ($20) + TS/MLV ($80) + SDcard ($10) = $1077

Less than $200 extra for a higher quality product.

James

[elaw]

Well put! And in fact I'd say $20 for assembly (I assume that refers to installation of the RTC and knock modules) is pretty conservative - I'm guessing most electronics techs would charge closer to $50 for that job.

I think what might account for part of the "sticker shock" for some people (and I count myself in that group) is that with the pro you're spending all the money at once, whereas going the other route you can upgrade incrementally. In my case, an MS2/v3.0 system came with a car I bought and spending to bring that to what I'm running now was in relatively small chunks spaced out in time. The tradeoff of course being that in the interim, I was running inferior hardware.

[ol boy]

Why not add a MS2 chip to the MS3 daughter board to make a "MS5", the MS2 chip would do the trans code while the MS3 handles all the fuel, ign...ect. Becomes a Ford EEC-V replacement ECU. Adding trans control will move MS up and beyond all other aftermarket systems. As we all know the trans control will work on all brands and types of automatics.

The Pixhawk autopilots have a Cortex M4 main processor and a M0 as a back up.
https://pixhawk.org/modules/pixhawk

A buddy of mine wants to use this hardware platform to do EFI and trans control all in one.

[Marek]

And while I have tremendous respect for the effort that goes into the development and support of the firmware, that's not really germane to this comparison because they both run the same firmware!

That can't be true:- the MS3pro looks to have native 12 cylinder support whilst the MS3/3X only achieves this in part. For example, if you power up a MS3/3X with 12 cylinders, only the first 8 cylinders get a priming pulse.

Look at the address lines from the processor and you'll see that they have subtly different connections. I don't think you can even use a knock module if you are a MS3/3X user with 12 cylinders as H1 and H2 are used for fuelling the extra cylinders.

At the limit, the two products can easily be differentiated.

kind regards
Marek

[dontz125]

Actually, the MS3-Pro has 12 fuel and 8 spark channels, plus the knock module. The boxed MS3/X has 12 each fuel and spark (the last 4+4 channels need something like the jbperf 4-channel driver card). The firmware allows you to specify which pins are used for fuel channels K and L, so you don't necessarily lose them IF you install the knock module. Note that if you go with an external knock module, you do NOT lose the last 4 spark channels. This is not an option with the MS3-Pro.

If you have "12 cylinder support" selected in the project properties and only 8 cyls prime, this is a bug and should be reported as such.

[muythaibxr]

That can't be true:- the MS3pro looks to have native 12 cylinder support whilst the MS3/3X only achieves this in part. For example, if you power up a MS3/3X with 12 cylinders, only the first 8 cylinders get a priming pulse.

It's a different build of the same firmware. Or in other words, it's the same base source code built with different flags to enable the ms3pro-specific features and I/O. We don't maintain a separate code base for the ms3pro. I would consider elaw's statement correct that it is the same firmware.

[dontz125]

Correction to my last; the selectable injector channels are I and J, *not* K and L. This only solves the problem for the RTC; if you use the internal knock module, you lose channel K (PM2 / H2), which also means you can't use channel L. My apologies for any confusion.

[jacky4566]

Bump. Has there been any update on the MS4?

My wish would be to make it cheaper. Yes I understand the costs of R&D and manufacturing but i feel the Bill of Material for the PCB could be cut in half with modern components and internalization of things like the USB bus.

For example, the current MS3 MC9S12XEP100 brain chip costs ~$22 just for the chip. Switching to an ARM4 chip like the MK22FN1M0VLQ12 would net the following:

• $4.71 per chip instead of $22
• 120Mhz versus 50Mhz
• Built in USB and Frequency oscillator further reducing costs
• Double ram for data logging
• More I/O 100 vs 92 with the given chip

[jsmcortina]

That chip doesn't seem to include the number and type of timers that are needed for the core engine control application.

James

[jacky4566]

I was only using particular chip as an example that in 5 years the world of RISC microprocessors has advanced quite a bit.

What are the requirements for timers in the MS project?

Is there any direction on MS4 yet? I just curious if we will see another MSPro like product or another DIY type unit.

[racingmini_mtl]

You also need a 5V CPU since all the automotive sensors are 5V (as far as I know). That makes using an ARM-based CPU quite difficult since there are very few 5V ones.

The other CPU types that are built specifically for engine control are not cheap and require (very) expensive development tools.

Jean

[Enki]

5v signals can be scaled for use on 3v CPUs. See it all the time on Arduino stuff (not saying such is comparable to MS3).

[dontz125]

That you CAN do something doesn't mean you should. You're talking about the signals that may or may not blow up your engine. It's one thing to scale ADC inputs on an Arduino, but it would be unprofessional from an engineering standpoint to do such a thing in a commercial product.

[jacky4566]

NXP's LPC line (and probably many others) has 5V-tolerant GPIOs OR level shifters are also common place in the instrumentation field as they migrate from 5v to 3.3v so I wouldn't rule out ARM all together.

My point is there must be a faster better stronger cpu than the one introduced in 2004.