Altium Designer new pricing model and high end / low end tool in development

[koko79]

Hey guys just caught this - http://www.altium.com/resources/investor_announcement/asx_releases/ASX_Announcement_Altium_Investor_Presentation_FY2014.pdf
Can't work out if this is good or bad for Altium!
Any thoughts?

[EEVblog]

Started the development of a new PCB Design product targeting the entry-level market

and

Restructure Altium Designer into a suite of software with different pricing

This is absolutely the best thing they can do, and it's what the customers (and many employees) have been screaming for for about a decade now. Seriously, a decade.
But I fully expect them to screw it up, as Altium always seem to do, but at least they will try.
I seriously need to do my video with my tips for how they should do this.
I heard it would be October, but don't quote me on that.

And they finally admit publicly:
"Over 90% of Altium's revenue comes from PCB business"
Which means that the 13 years and the entire company fortune they spent on FPGA has been a complete waste.

[koko79]

Well I think something needs to happen. They recently released financial figures year ending June 30 2013. US $69.2 million in sales which resulted in US $1.4 million in profit after tax. Hmm interesting times ahead!

[EEVblog]

Well I think something needs to happen. They recently released financial figures year ending June 30 2013. US $69.2 million in sales which resulted in US $1.4 million in profit after tax. Hmm interesting times ahead!

I should have bought those shares at 10 cents...

[koko79]

8AU cents dividend for those that did! Not sure if that eats into the US$1.4m profit after tax.
Loos like they want a piece of the Cadence, Mentor and Zuken high end PCB pie.
Is it possible to write that higher end software and get it to market on those profit margins??
Those high end customers will still want a SI solution.

[AlfBaz]

Those high end customers will still want a SI solution.

Whilst their current offering in this area is sorely lacking, it wouldn't take much to make it competitive.
They already have access to a 3d model of the copper they just need to get their hands on a field solver engine.

On the layout side they need to sort out some of the length matching stuff, like being able to recognize an entire signal path regardless of series terminators for example

[EEVblog]

8AU cents dividend for those that did! Not sure if that eats into the US$1.4m profit after tax.
Loos like they want a piece of the Cadence, Mentor and Zuken high end PCB pie.
Is it possible to write that higher end software and get it to market on those profit margins??

Not originally, but the majority of the (expensive) original PCB code group are now gone.
I have no idea how competent the new cheaper hires in Shanghai and the Ukraine are, but assuming they have a big enough group who know their stuff I don't think it would be that hard to get the SI up to scratch. Not really super duper high end, but good enough to compete for most tasks, and that's all they need. To shoot for industry leading SI would be silly, just like FPGA was.

[mikeselectricstuff]

Altium could release a powerful, easy to use, quick to learn PCB tool tomorrow, with no development cost.
It's called P-CAD2006
 

[EEVblog]

Altium could release a powerful, easy to use, quick to learn PCB tool tomorrow, with no development cost.
It's called P-CAD2006

Sure, or Circuitmaker 2000, but either would be a bad move.
The code base is no longer supported, and it doesn't provide seamless upgrades to high priced version of Altium Designer.
They would be crazy to do anything but release stripped down Altium Designer, which has never actually been that hard to do if thy wanted to. But sounds like they are going the whole hog and rejigging the entire code base to separate the functional modules etc.

What they need is:
- A free version of some sort. This is absolutely essential for complete market domination of Altium designer. But sadly is something they are very likely to overlook. Eagle having a free version is the sole reason why it is now the defacto standard PCB tool in the hobbyist/hacker/maker community. Eagle has more seats than Altium, and they could steal a majority of those seats within months with a good free version. A low cost paid version will not cut the mustard here. If they want total market domination they must do this.
A free version would have absolutely no company support, and could even have a non-commercial license like Eagle does. But they could one-up Eagle by having some limited commercial use for free.

- A low cost version which that are obviously going ahead with. There are many ways to skin this cat. To be commercially viable I suspect the price has to be somewhere in the $200-$500 category. They could perhaps even have a 3 or 4 tear pricing system. e.g.
1) FREE
2) sub $100 limited commercial use
3) $200-$500 sweet spot for the low end users.
4) Maybe $500-$2K for slightly higher end users
5) Full version as it stands currently

For the The low cost cost versions, you wouldn't bother with any company sales or support at all, it would be fully automated PayPal ordering/download and forum support only.

But here I think is the kicker, all versions must include at least the following:
- 3D
- BOM/Supplier/price stuff
- Full library access
- Full schematic

Without those things, you aren't going to get the "wow" factor that will draw people into the Altium world.
So they need to differentiate the product based on other things, like for example:
- FREE and limited version get say 2 layers and size restriction.
- The "low cost" $200-$500 price point gets unlimited size and 4 layers.
- The high end or full versions get stuff like interactive routing, autorouting, simulation, SI, FPGA, pin swapping, embedded, snippets etc etc.

You could argue that manual interactive routing belongs in the essential "wow" category.

And really, Altium, no one gives a shit about FPGA and embedded stuff, really, they don't, they never have and they never will want Altium to provide that solution. Pin swapping if the only FPGA thing that matters. You will never be able to have a real industry solution in FPGA or embedded, ever. You spent 13 years on that dream, you failed, drop it and move on, leave it to the FPGA manufacturers and companies that do it right. But you know that of course, and now admit that 90%+ of your income comes from PCB, as it always has and always will.

I do think that around $500 is the upper sweet spot to take on the low end market. $1K will be too much, and under $200 is chump change.
If this low cost version doesn't include any company sales or support effort as I mentioned then it becomes money for jam. That is the position you want to be in.

[IanJ]

I'm looking forward to having Altium as an affordable alternative to Eagle........but also having Eagle raise their game to counteract..................well, I live in hope anyways!

Right now, Altium is too expensive........and Eagle won't listen and fix some fundemental issues (form me) with their product (I just have the hobbyist license). Shame!.....because the money is waiting.

Ian.

[alexmackuk]

This has probably already been said/asked (but I just signed up so forgive me! LOL)... but the thing I find hard is comparing the packages available in any meaningful way. I currently use something called Proteus (from Yorkshire - formerly ISIS and ARES a separate schematic and PCB pair but now an integrated tool) which is pretty good (probably low- to lower-mid-end) for anything up to 6-layers but lacks finesse when dealing with things like BGA packages and multi-layer tracking. For instance it lacks anything to deal with tracks that need to be length or impedance matched such as LVDS or tracks that you would want routing together (such as data or address lines) - anything like that has to be manually tracked.

I've always vaguely lusted after Altium Designer without really being able to justify the cost so would probably bite at the cheap version... but I would not really know what benefits I would be getting until I actually tried it!

Have you ever done any detailed comparisons of low-, mid- and high-end (although they're all out my price range!) design systems? You can find reviews, etc on the web and you can watch the sellers videos and blurb but the reviews are often partisan and the promotional materials only every show you the good stuff and not the bad  .

[mikeselectricstuff]

In terms of limits, I think maybe a component count limit may be an alternative to a size limit - not sure of the pros/cons but it is something they did with PCAD (6 layer, 400 components I think).

I'm not convinced about the whole commercial/noncommercial thing as it's completely unenforceable - I think you have to give some functionality benefit for buying the lowest end  - maybe an increase in size/component count. Maybe 3D.

[ElectroIrradiator]

Am I the only one, who feel that 4 layer support even for a free version seem almost mandatory these days?

With 3+ layers you can have an uninterrupted signal path on one layer, located above an unbroken ground plane. This leaves the other side plus any remaining layers for routing power rails and control signals. Given the speed that many hobbyist widgets run at these days, then even very small items can frequently benefit from proper RF layout techniques and shielding. I would almost say it is a bit of a 'crime' to force hobbyists and beginners to only use two layers, as in many instances it would forces them into making various suboptimal kludges on high speed analog and digital designs. A two layer limit would prevent them from learning how to do things properly, nor could you teach them how due to lack of proper tools

With only two layers you are forced to reserve one of them for the ground plane, which means you will often need jumpers in order to make 'clean' RF technique aware designs for the signal path(s). Or is there some deep PCB layout topological secret regarding two layer RF designs, which no-one has bothered to teach me yet?

[EEVblog]

In terms of limits, I think maybe a component count limit may be an alternative to a size limit - not sure of the pros/cons but it is something they did with PCAD (6 layer, 400 components I think).

Sure, either way would work if implemented sensibly.

I'm not convinced about the whole commercial/noncommercial thing as it's completely unenforceable

It's not, and that's not the idea anyway.
If the "commercial" tool is priced resonably enough, a good majority of users will go legit.
That is why I suggested the Eagle method of $569 for the "commercial" version of the same stripped free down version. It's a no-brainer price for almost anyone.
Then you have the proper intended "low cost" version at the $200-$500 price point.

- I think you have to give some functionality benefit for buying the lowest end  - maybe an increase in size/component count. Maybe 3D.

I believe 3D must be present in all version, even the free one. It's the "killer feature" that will suck people into the tool. Although they could leave out mechanical 3D DRC or something like that.

[EEVblog]

Am I the only one, who feel that 4 layer support even for a free version seem almost mandatory these days?

I don't think so.
Many quite complex designs can be fitted on a 2 layer board.

With 3+ layers you can have an uninterrupted signal path on one layer, located above an unbroken ground plane. This leaves the other side plus any remaining layers for routing power rails and control signals. Given the speed that many hobbyist widgets run at these days, then even very small items can frequently benefit from proper RF layout techniques and shielding. I would almost say it is a bit of a 'crime' to force hobbyists and beginners to only use two layers, as in many instances it would forces them into making various suboptimal kludges on high speed analog and digital designs. A two layer limit would prevent them from learning how to do things properly, nor could you teach them how due to lack of proper tools

True. But most hobbyists are likely going to squeeze designs onto 2 layers anyway to take advantage of the enormous range of low cost 2 layer PCB proto services.

[EEVblog]

Have you ever done any detailed comparisons of low-, mid- and high-end (although they're all out my price range!) design systems? You can find reviews, etc on the web and you can watch the sellers videos and blurb but the reviews are often partisan and the promotional materials only every show you the good stuff and not the bad  .

No, I haven't done such videos. It's hard to do that when I've been using Altium/Protel for more than 20 years. It wouldn't be a fair comparison when I'm intimately familiar with one package, and not the others.

[alexmackuk]

Yes, I'm in much the same boat with Proteus - I've been using it for long enough that things that it probably does wrong are masked by my experience and other packages seem clunky by virtue of unfamiliarity .

It was so much easier to choose when it was sticky crepe traces and pre-cut pads at 2:1 or 4:1 on mylar sheets! (I've still got the A0 light-box  ).

If Altium put something out at <£500 then it will be a no brainer to give it a proper try.

[olepr01@gmail.com]

As an example of someone elses take on the problem of expensive software/poor hobbyists, look at Mathematica from Worlfram. The full package costs an arm and a leg (an extra leg if you need to pay VAT), but rather than releasing a crippled version they sell the complete, full-featured program for home use at a massive discount. The only difference is that the software is only licenced for noncommercial use. Those who want to stay legit use it within the legal limit. Those who do not care about the legitimacy of the software they use would probably just pirate it anyway, but this way at least theres some revenue from them. So theres no real reason to have a crippled version, IMHO..

In my opinion the right way to go, I can at least say that Wolfram has got more of my personal money than any other software-vendor.

[graynomad]

Yeah I know, slightly old thread but anyway I have two things to add.

2/4 layer, with the cheap prices of PCB fab these days I never even consider 2 layer any more, 4 layer is much easier to route.

And from the "I heard it from a mate" department, just the other day I heard from a mate who works for a large Oz company that the light version will NOT include schematic capture. Personally I find that impossible to believe and said as much. My mate felt the same and also said as much to the rep (or whoever) he was dealing with. But that was the story.

I still can't believe that because how on earth would it work without importing the schematic?

I used to use pre-release Protel versions (sort of beta testing I guess) when it was written by two blokes in a Tasmanian shed, then I left the game for a long time and when I came back Protel had morphed into humungous bloatware of which I only need about 20%, if that.

Anyway it will be interesting to see what happens with this.


BTW I reckon Dave's feature/pricing structure is pretty spot on.

EDIT: I see from that marketing PDF

Breaking Altium Designer into a suite of software products

Maybe they are going to split the F(ing)PGA and other stuff from the core schem/PCB software.

______
Rob

[EEVblog]

And from the "I heard it from a mate" department, just the other day I heard from a mate who works for a large Oz company that the light version will NOT include schematic capture. Personally I find that impossible to believe and said as much. My mate felt the same and also said as much to the rep (or whoever) he was dealing with. But that was the story.
I still can't believe that because how on earth would it work without importing the schematic?

That's what Altium are famous for, doing the ridiculous. I hope that's not still the case now that Nick Martin is gone.

[poorchava]

Ok, I'm lost here... How the hell are you supposed to use PCB CAD without schematic capture? Write the netlist by hand to a txt file?!

[graynomad]

Got me buggered, maybe import from Eagle

It's hearsay though, not verified and one would assume almost impossible for even Altium to shoot themselves in the foot that badly.

______
Rob

[andtfoot]

Ok, I'm lost here... How the hell are you supposed to use PCB CAD without schematic capture? Write the netlist by hand to a txt file?!

I've done it before for really simple stuff. Just drag out the footprints only and connect them based on relevant chicken-scratching/back-of-napkin circuit.

[EEVblog]

It's hearsay though, not verified and one would assume almost impossible for even Altium to shoot themselves in the foot that badly.

Don't put it past them, seriously, they once made the PCB tool optional extra. As in, you got everything BUT the PCB tool.
They marketed it under the slogan "turning the world of electronics design upside down"

[IanJ]

Ok, I'm lost here... How the hell are you supposed to use PCB CAD without schematic capture? Write the netlist by hand to a txt file?!

I've done it before for really simple stuff. Just drag out the footprints only and connect them based on relevant chicken-scratching/back-of-napkin circuit.

Oh yes, just like Wintek's SmartWork back in the 80's............so maybe Altium have aquired the source code from them........ 
I better brush up then....F2 was place-pad wasn't it?

Ian.