EEVblog #527 - Altium Entry Level PCB Tool Rant

[walshms]

Yes, that's the tough part for many.  It's a mindset that needs to be overcome, and it's very hard for a lot of the old-school business types to think in terms of winning through volume instead of higher margins.

Just remember, this isn't a race to the bottom.

No argument there.  I agree.  And you've hit the nail right on the head here, I think:

What it is is actually a crawl to the top, by virtue of being a major player at bottom. It's all about enticing customers up the food chain into your higher end products.

This is why I was saying that it needed to be affordable to remove the restrictions they might place on a "low end" product.  I'm happy to pay for what I need, provided it's reasonable, and I expect that most people are as well.

Altium now know this.

A very good thing.  Even better if they take these messages to heart.

So they will be looking well beyond that, and at the bigger picture of general market domination.

As well they should.  I'm not suggesting they should be a charity -- quite the contrary.  I'm suggesting that they should be smart about it, and hopefully, they get that message.  Should really be a cakewalk if they do it right; they can rely on community support rather than having to dedicate support resources for it, for example.  But there's payback as well -- component libraries developed by the community can flow back to them. 

It can be a win-win, and should be.  And I think it's safe to say that if they took that approach, they'd find people very willing to help them too.

[walshms]

That's got to be hugely expensive.

Yes, it was well into the 4 figures for a couple of boards

Ouch!  That hurt just reading it!

Yes, prototypes, which were a slightly inferior quality than their usual production stuff, presumably because they rush it and take much less care than usual.

Hate it when that happens.  Especially when you're paying top dollar.

It becomes the norm when no one cares how much money you spend.

I suppose it's okay if you have money to burn... but I think I can relate to that.  I had a customer who took that approach, and wanted a video camera set up to capture his pet project -- two wind generators that were being installed on their property -- to be streamed live to the 'net 24x7.  He didn't care what it cost, he just wanted it done, and done yesterday.  Wound up spending more than $10K before it was realized that the connectivity to the site was so poor you couldn't get streaming video over the link at better than 1FPS.

At which point, of course, they signed an agreement with the local cable operator to bring in fiber on a peering agreement.   

So what should have been a no-brainer became a $2.4K per month for five years deal for the local cable operator and a $10K waste because immediately after it was up and running he lost interest and decided to pull the plug.

And these are the people that wind up in executive positions.  Go figure... no matter how hard I try, I can't.

[jancumps]

...
•   Mechanical library creator - 3D editor – in stp format and supporting 3D printers (makerbot, RepRap,…) to prototype small plastics

...

Can you drop this particular suggestion. They've just started to focus on their core business again.

[David_AVD]

Yep, I'd rather a cheaper version with just schematic and PCB.

[mikeselectricstuff]

It's also not uncommon for me to do PCBs on a same-day turnround, which isn't possible for 4L due to the lamination process.

We used to get 8 layers with all the bells and whistles regularly turned around in 24 hours in Taiwan.
Cost a butt load of course, but possible. Some poor bastard works all night on it and they dedicate a line to it.
With Fedex priority/super duper emergency service you could have the thing in your hands in 2 days depending upon what hour you ordered it.

As a contrast, I've had same day 2L PCBs (files in by 10:30, shipped out same day, or I could have collected 5PM) from my local place, and only cost about GBP220 for two A5-size panels.
 

[EEVblog]

As a contrast, I've had same day 2L PCBs (files in by 10:30, shipped out same day, or I could have collected 5PM) from my local place, and only cost about GBP220 for two A5-size panels.

Yes, we still have a couple of same day turn PCB houses in Oz. They will also do 8+ layers as well, but usually 2-3 days or something at best. I guess no one here wants to work 24 hours straight!
So for us it was quicker to get 8L 24 hours turn from Taiwan and have it Fedexed.

[Lukas]

Can some Altium user give an overview which features make Altium infinitely better than Eagle?
Does Altium have an instantaneous forward annotation like Eagle? For small projects I'm used to do schematic and board concurrently.

[elgonzo]

Just remember, this isn't a race to the bottom. Altium will not win anything much on volume alone here. What it is is actually a crawl to the top, by virtue of being a major player at bottom. It's all about enticing customers up the food chain into your higher end products. To do that you need large volume (but not necessarily large income) in the bottom end. Because it all comes down to statistics. The more customers you have at the bottom end, the more on average you can entire to your higher end products.
Altium now know this.

Even if you make $100 profit on a $100 low end product (possible if you do it right and fully automate it, and assume your high end pays for the little upkeep required), that's still not serious income. For Altium to double their yearly income of $69M at that level, they would have to sell 690,000 seats. Even a massive figure like 100,000 seats would only bring them in a lousy $10M extra income, which isn't much for a company like Altium.
So they will be looking well beyond that, and at the bigger picture of general market domination.

Interesting...

Assuming that the average EE engineer, who will become a possible AD pro user, has to take certain paths through education, wouldn't it be better to bombard these paths with educational or free licenses?

Or is Altium doing it, because for smaller professional design houses and fabs, Altium is already too much of a cost factor? If that is not the case, then you would look at semi-professionals and hobbyists when going to offer a low-cost version (please correct me if i am wrong). How would a statistically relevant amount of these individuals end up in a company as a professional AD users (or having any impact on a company's decision to purchase whatever software solution)?

Or is it that Altium doesn't receive trust, with their commercial customers lacking the confidence in Altium to deliver working solutions, and Altium tries to improve how they are perceived?

I am under the impressen that Altium already dominates "their" market. If not, who would be the competition for Altium Designer? Cadence OrCAD, Cadstar?

[ddavidebor]

If altinium do a cheap and stable student version, my school will likely but it for 150pc

[EEVblog]

Assuming that the average EE engineer, who will become a possible AD pro user, has to take certain paths through education, wouldn't it be better to bombard these paths with educational or free licenses?

They have already been doing that since the start of the company back in the 80's. Perhaps not as aggressively in recent times.

Or is Altium doing it, because for smaller professional design houses and fabs, Altium is already too much of a cost factor?
If that is not the case, then you would look at semi-professionals and hobbyists when going to offer a low-cost version (please correct me if i am wrong). How would a statistically relevant amount of these individuals end up in a company as a professional AD users (or having any impact on a company's decision to purchase whatever software solution)?

Altium is going after anyone who currently uses Eagle or some other low priced package. The want to dominate a 3 tiers of the market.

Or is it that Altium doesn't receive trust, with their commercial customers lacking the confidence in Altium to deliver working solutions, and Altium tries to improve how they are perceived?

Yes. Altium have one of the worst reputations in the industry of ignoring what their customers want, and pushing into stupid new areas that were doomed to fail. The only reason customers didn't leave in droves after the last decade is because Altium essentially didn't have a rival in the same price and feature/usability bracket.

I am under the impressen that Altium already dominates "their" market. If not, who would be the competition for Altium Designer? Cadence OrCAD, Cadstar?

It's a hard one to answer. Yes, you could argue that Altium dominates in the mid level $2K-$10K market.
But Altium isn't as complete a solution as the top end Cadence and Mentor packages, so on some scales doesn't really compete there. But traditionally, Altium have always treated the "big two" as their main rivals. Now they have had another "oh shit" moment, and realised they better take the low end Eagle market seriously.

[JoannaK]

I am under the impressen that Altium already dominates "their" market. If not, who would be the competition for Altium Designer? Cadence OrCAD, Cadstar?

It's a hard one to answer. Yes, you could argue that Altium dominates in the mid level $2K-$10K market.
But Altium isn't as complete a solution as the top end Cadence and Mentor packages, so on some scales doesn't really compete there. But traditionally, Altium have always treated the "big two" as their main rivals. Now they have had another "oh shit" moment, and realised they better take the low end Eagle market seriously.

Has anyone any real numbers.. I know the companies themselves like to brag a lot about 'seats' or other metrics, but IMHO it's quite obvious those numbers are more/less marketting fluff, since it's quite obvious not all of them can be market leaders.

Would be really nice to see what the market shares really are. Someone like http://www.garysmitheda.com/ apparenly has at least some data, but the pricing is so astronimical that I'd rather wait for Reader's Digest edition ..

(edit.. quotes fix, plu s couple typos)

[Carrington]

For anyone who wants to make the leap from Eagle to Altium.
http://altiumdesignerblog.blogspot.com.es/2011/07/how-to-import-eagle-files-into-altium.html
And attached my version: "export-protelpcb_v0.2.ulp" For now it only works in version 5.x of Cadsoft Eagle.
"Enjoy it"[/s]
Oh men:
http://www.altium.com/en/video-eagle-importer

[Rigby]

But, when it comes to this 1000$ piece of code you want for your hobby, that will be used for years and is cheaper in the long run than the other leisure activities ( total cost over the amount of time used ) oh no. it's gotta be free , or less than 100$...

That is not at all the argument.  LOTS of hobbyists pay dearly for their hobbies, and that includes electronics.  I purchased Visual Studio for one of my hobbies, Photoshop for another, and another requires that I buy expensive, hard to acquire hardware any time I want to do anything.  It's not that hobbies (or electronics) should be free, it's that Altium NEED to release a free version if (and only if) they want to truly compete in the hobbyist market.  That's all.  There are already lots of free options in that realm, and if Altium wish to compete in that market, then they need to provide a product that fits in the market.  One cannot compete in a market unless one prices their semi-equivalent product similarly to other products in that market.  Once you're established a bit, you can start shoehorning out your competitors.

Yes, there will always be free software folks who think everything should be free and open and all that, but they are a very vocal, very small minority. 

[Rigby]

While there is nothing inherently disastrous about a good (!) online tool

Yes there is - your workflow becomes entirely dependent on a fast, reliable internet connection and servers at the other end. Either falls over and you are completely screwed and absolutely nothing you can do about it.
I really can't see why anyone would think online design tools are a good idea. File-sharing, collaboration - fine, but having to be online during an actual design/layout process, hell no.

Meh, technically that's not exactly true.  Web apps can store loads (megabytes) of stuff locally without relying on an internet connection, and there's no requirement for a "fast" internet connection (by my definition); you just need to be connected when the app loads and whenever it decides to reach out.  It's all up to the app developer on when that happens.

I agree though, that there is no reason at all for this type of tool.  There is no valid reason the make a web-based design tool, and there's not a customer in the world who knows what they are doing that would ever ask for this, which makes me question the motivation for the creation of such tools.

[Bored@Work]

Meh, technically that's not exactly true.  Web apps can store loads (megabytes) of stuff locally without relying on an internet connection, and there's no requirement for a "fast" internet connection (by my definition); you just need to be connected when the app loads and whenever it decides to reach out.  It's all up to the app developer on when that happens.

You mean HTML5 app caching? A few month ago I experimented with it. It doesn't cut it, except maybe in one environment, where it is almost bearable - Apple iPad/iPhone. But even there, like in all the other environments, it is basically a fragile, difficult to manage technology. The browsers I tried behaved differently, even if it was the same browser but on a different platform. Some came up with scary messages, deterring the user from granting the necessary rights to do the caching. Others had strange ideas when to cache and when not, despite the same manifest. Some were extremely protective of the cache, making it very hard to update a web app. Others dropped the cache at the blink of an eye for no particular reason.

which makes me question the motivation for the creation of such tools.

It is some mental deficiency those web "programmers" have. Fanaticism, stupidity, not knowing better, selective vision. Everything not web based and not running in a browser is considered old junk by them.

[yym]

the free version (or all of them) should be more like protel99, remove all that fpga crap, keep it simple, just schematic and pcb, size/layer restrictions are acceptable, reazonable size, 2 layers, better autorouter, that acually works with only 2 layers. Keep ui simple as possible (protel99) handy shortcuts, not fancy and slow gui frameworks.

[hikariuk]

It is some mental deficiency those web "programmers" have. Fanaticism, stupidity, not knowing better, selective vision. Everything not web based and not running in a browser is considered old junk by them.

I'm a horrible person and tend to look at web developers and go "Aww, they think they're programmers.  Isn't that cute?"  I have an inherent distrust of anything that requires me to run a web browser and be online to run.  It's the main thing I don't like about mbed.  I like having all my development tools sat on my computer in front of me; not on some server at some other place in the world.  Oh and screw all this cloud bollocks too - reinventing existing ideas badly.

[FrankBuss]

I have an inherent distrust of anything that requires me to run a web browser and be online to run.  It's the main thing I don't like about mbed.  I like having all my development tools sat on my computer in front of me; not on some server at some other place in the world.

Recently mbed was open sourced and all its components, like CMSIS (BSD licence). There was no no problem when I cloned the embed github repository and compiled my programs locally with the integrated Python build framework, it's all documented. Works good with the free ARM GCC compiler.

I started to port the FPGALink framework to the LPC controller, here is a short manual how to setup the mbed framework for Windows and how to compile the FPGLink firmware with it, but can be used for your own projects as well.

[Agent24]

If there was a free version like Eagle Light then I would probably have at least tried it by now.

Eagle Light was the first PCB software I tried because it was free, but I never liked it.

I discovered KiCad later on and found it to be a no-brainer for me.
Much easier to use, free, open source, no artificial restrictions and cross-platform. Sure it doesn't have some advanced things but as a hobbyist it does what I want and I love it a lot more than Eagle.

Currently I see no reason to switch from KiCad to anything else.

[poorchava]

Can some Altium user give an overview which features make Altium infinitely better than Eagle?
Does Altium have an instantaneous forward annotation like Eagle? For small projects I'm used to do schematic and board concurrently.

-It's not Eagle
-It's not Eagle
-More detailed DRC
-actually working and practical library management
-3d mode
-everything is hotkeyed and can be edited at will
-online DRC
-interactive routing with hug, push and walk-around modes
-schematic snippets
-better interface
-teardrops
-via stitching
-works well with pcb of unusual shape
-making new components is much easier and faster
-complete, easy and fast to use fabrication output tools

Aside from somewhat lacking high-speed tools (which I almost don't use anyway) it has all you need for completing any design from a led blinker to 50cmx50cm 16 layer boards.

[Agent24]

-It's not Eagle

-3d mode

-online DRC

You just made me remember what I hate most about Eagle: no online DRC!
No realtime 3D view is also a very big limitation and annoyance.

[mikeselectricstuff]

-making new components is much easier and faster

Any company that makes PCB software where component generation is a pain to do has no clue. Libraries will never, ever be complete, so it's something every user needs to do regularly.

[VEGETA]

OK, I am very excited about this Altium free software version... But, when will it be released? is there any place or post to know about release data or should we wait forever to get it.

I haven't seen any single news post on their site! I started to doubt they take it seriously or not.

[EEVblog]

OK, I am very excited about this Altium free software version... But, when will it be released? is there any place or post to know about release data or should we wait forever to get it.

Just rumors really. Some Altium sales people have reportedly said October.

[VEGETA]

OK, I am very excited about this Altium free software version... But, when will it be released? is there any place or post to know about release data or should we wait forever to get it.

Just rumors really. Some Altium sales people have reportedly said October.

well, it's October already ^_^

I think I will wait.