Looking to get into S-100 Computing - Buy or Build?

[ArchieX]

Hello everyone! This may be a long-winded post...

I got bit by the vintage computing bug long ago when I was young due to a mutual interest in anything retro or vintage, and my interest in computers. As such I have a whole rec room filled with VIC20's and C64's, 8088's through Pentium-era systems and parts, and a small horde of era-related software that I haven't played with in forever. Mainly because I spent so much time playing with it years ago.

But I always wanted to go older - an 8080 system - but never got a chance due to the absolute rarity of them. I've never even seen one in real life. But then Grant Stockly came by with his detailed Altair 8800 replica and I realized I might have a chance! But naturally as things go with me, I never had the cash at the time to purchase one. I've been desperately holding out for the chance that he would start producing them again since '07 but doesn't seem like it will happen.

Now there's no way I'm going to drop however many thousands of dollars for a genuine Altair 8800 that pops up every now and then on eBay. Even my strong desire to play with one has limits, but then I recently started thinking, "why not build my own"?

I stumbled across the plentiful resource of S100Computers.com and only recently realized that there is a small community building new cards for S-100 systems, meaning there actually is a way to build one from scratch (figuratively speaking) and I didn't have to wait until my hair turned grey to own an 8080-based system!

But as much as I want to build one, I have to admit that I have never done anything like this before. I've never laid hands on hardware like this, I've never debugged physical cards or circuits, and I've never even soldered electronics other than battery cables for my pickup truck. I do have some experience in building simple circuits in robotics class and assembly programming way back in college ten years ago, but haven't retained much of what I had learned.

I realize that taking on a project like this will involve an intensive study not only in soldering and wiring, but also the intricacies of the associated circuit designs, programming, and oh-so-much more. My end goal is to eventually learn how to design and fabricate my own cards, and potentially even my own software.

So I'd like to ask the more experienced members of this forum, is this a feasible project for someone with my limited experience? How many of you have built your own S-100 system? How difficult (in hindsight) was it to get it operational? Would you have rather bought an existing system or still build your own?

 

[MarsMan2020]

Hello everyone! This may be a long-winded post...

I got bit by the vintage computing bug long ago when I was young due to a mutual interest in anything retro or vintage, and my interest in computers. As such I have a whole rec room filled with VIC20's and C64's, 8088's through Pentium-era systems and parts, and a small horde of era-related software that I haven't played with in forever. Mainly because I spent so much time playing with it years ago.

But I always wanted to go older - an 8080 system - but never got a chance due to the absolute rarity of them. I've never even seen one in real life. But then Grant Stockly came by with his detailed Altair 8800 replica and I realized I might have a chance! But naturally as things go with me, I never had the cash at the time to purchase one. I've been desperately holding out for the chance that he would start producing them again since '07 but doesn't seem like it will happen.

Now there's no way I'm going to drop however many thousands of dollars for a genuine Altair 8800 that pops up every now and then on eBay. Even my strong desire to play with one has limits, but then I recently started thinking, "why not build my own"?

I stumbled across the plentiful resource of S100Computers.com and only recently realized that there is a small community building new cards for S-100 systems, meaning there actually is a way to build one from scratch (figuratively speaking) and I didn't have to wait until my hair turned grey to own an 8080-based system!

But as much as I want to build one, I have to admit that I have never done anything like this before. I've never laid hands on hardware like this, I've never debugged physical cards or circuits, and I've never even soldered electronics other than battery cables for my pickup truck. I do have some experience in building simple circuits in robotics class and assembly programming way back in college ten years ago, but haven't retained much of what I had learned.

I realize that taking on a project like this will involve an intensive study not only in soldering and wiring, but also the intricacies of the associated circuit designs, programming, and oh-so-much more. My end goal is to eventually learn how to design and fabricate my own cards, and potentially even my own software.

So I'd like to ask the more experienced members of this forum, is this a feasible project for someone with my limited experience? How many of you have built your own S-100 system? How difficult (in hindsight) was it to get it operational? Would you have rather bought an existing system or still build your own?

I think it's totally doable, the new S-100 Z80 Single Board Computer that John M. recently designed is a great starter board - you can even bypass the regulator and feed it 5V without any S-100 backplane and still get started running CP/M, then drop the card into a backplane later and use it with the other S-100 boards.

Through hole soldering is pretty easy once you get some practice. The thing that made the biggest difference for me was investing in a nice Weller digital temperature controlled iron. It is sooooooo much better than the el cheapo ones. A fresh spool of good quality solder, a wick to clean up any mistakes, and away you go. Keep the tip and the parts clean is the most important thing technique wise.

If you'd like something smaller with less chips to get started, I started out with the Zeta which is a true SBC (no bus) to gain confidence before building anything larger. You can add the piggy-back ParPortProp board to the Zeta to make a nice 2-board stack with built-in VGA, PS/2 Keyboard, and SD Card if you want.

 

[MarsMan2020]

Also - there were a lot of S-100 systems after the Altair and if you post around the forums here, sometimes people will help a new person get started with some hardware. A box that needs some TLC, things like that.

There is one person eBay who has been driving all the S-100 stuff up - not representative of people in the hobby who tend to be pretty friendly.

 

[MicrocomputerSolutions]

If you can't even solder yet, I probably would not learn how to on computer boards. Soldering is a technique that looks easy when someone experienced is doing the soldering, and is easily learned if you are serious about learning.

But you're better off starting out on something easier/less complicated. There are still some companies offering kits for electronic items that you can use like clocks and radios. There are also a fair number of unbuilt Heathkits still floating around,but the prices have been rising since Heathkit went out of business (late 1980s?).

I would not recommend starting out with a pile early, poorly designed, and poorly already built S-100 boards either. Maybe start with running, factory built boards that were designed in the early 1980s. Later, when you get your feet on the ground. have gained some experience, and have figured out what part of the S-100 computer interests you, you can upgrade your systems with more features or modre advanced features, or if you like, modern boards.

Unless you a pick an S-100 Product line that was sold as a complete solutions (mainframe, disk enclosure, system boards, drives, and software) you are going to need to be an Electronic Tech to fix the boards you buy, and a Software Programmer to customize the software (including the operating system) so your computer will boot up, and be able to run programs.

There are companies starting from the beginning that sold complete solutions, and companies at the end of the S-100 era that offered some very advanced (for the time) turnkey systems.

I starting building my first S-100 System back around 1976-1977. I didn't get it running until close to 1980. My first board set was from SD Systems. SBC-200 4mhz Z-80A processor board with serial and parallel port, Expandoram II with 64K, Versafloppy II floppy controller. I bought a Integrand Research Model 800 mainframe with 20-slot motherboard, and two Shugart SA-400. With a Heathkit H-19 terminal that was over $2,000.

I still have this board set and recently bought a newer model Integrand Mainframe as a new home for my original SD Systems board set (I sold my original Integrand 800 mainframe about 25-30 years ago). I was an authorized Integrand Dealer back in the 1980s and 1990s. I believe that Integrand went out of business in the early 2000s.

You could probably get the same three SD Systems board set in used condition for less than $300 now (I paid about $1,000 with the SD Monitor PROM). SD Systems did come out with their own Serial/Parallel I/O board, and a Hard Disk Controller, that you could probably get from a eBay Seller. These boards are scare as they were made in smaller number than the basic three board set.

I went away from SD Systems because while they offered a complete solution (they sold complete turnkey desktops, pre-configured copies of CPM 2.2 and what they called SDOS, which was CPM 3.0), their boards did not play well with all other brands of boards because they were designed before the IEEE-696 Industry Standard was published. Compupro was building boards more to the IEEE-696 Standard, and the President of Compupro Mark G. was a member of the IEEE-696 Standards Committee, and the author of a very well receive book on the S-100 Bus. Compupro was building a complete series of compatible boards, S-100 mainframes, disk enclosures, and Licensed Digital Research Operating Systems, customized to run all the Compupro boards, which they sold with commented source code.

My 2nd complete computer (and 3rd, 4th, 5th, etc...) were all Compupro based, and I learned to modified the critical BIOS for different Compupro OS by readiing the commented source code from Compupro.

A few years later, I founded a company that sold and serviced thousands of individual Compupro board and hundreds of Compupro Systems. I still have parts of that first Compupro system running. Over the years as new boards, and newer revisions of the boards were introduced, I keep updating my original system.

If you want to look for S-100 boards, eBay is a place that you can see many different boards. If you see something that interests you, before placing a bid, inquire here about the item/s you're looking at, and people can tell you something about them. Same thing with the Sellers on eBay, we will be glad to tell you who you should stay away from. There is someone on eBay who is singlehandedly buying up all the low priced vintage computers on eBay (not just S-100), then flipping them back at ridiculous prices with fake stories about where they came from, and claiming that he is the original owner and has owned the stuff since it was new.

 

[tma]

Greetings,

I have been heavily into electronics building for over five decades. I started out building my test equipment from kits. I built a half dozen or more Eico, Heathkit and Paco instruments back in the day. In recent years I have designed surface mount PC boards and had them manufactured in China. I don't wish to discourage anyone from getting into this very interesting hobby but I do agree with one of the other posts with regards starting out with something a bit simpler to get some experience with soldering and working with circuits. Trouble shooting computer boards can be extremely difficult especially if you don't have the right instrumentation available.

I built my IMSAI S100 system from a kit in the late 70s. I started out with paper tape non-volatile storage, moved on to cassette tape and then finally 8" floppy disks and CP/M. This progression took me about two years even though I already had a solid background in electronics. There was a need to learn assembler programming along the way to get it all going. The learning curve for machine language can be quite large and can be quite difficult if you don't understand the basics. An S100 machine with a front panel can be quite helpful as it permits one to view the buses and control lines whilst single stepping through the code. An inexpensive alternative way to get started with assembler programming would be to play with some microcontrollers such as the simple 8 bit Microchip or Atmel devices. There are many inexpensive microcontroller boards available. The Arduino Uno board starter kit that uses a low end 8 bit Atmel controller might be worth considering. The kits typically cost less than $20 and one could experiment with building simple peripheral boards for it to control to gain some basic experience. There are inexpensive bare prototype boards available that will plug directly on to an Arduino board.

http://www.ebay.ca/itm/Starter-Kit-...839708?hash=item25b5db531c:g:agAAAOSwa-dWrn~a

With regards to soldering tools after over 5 years of use I am happy that I purchased an inexpensive combination temperature controlled pencil and hot air station. There are various eBay vendors selling slightly different versions of similar stations which typically cost a bit less than $100.00. There is an example here:


http://www.ebay.ca/itm/853D-3in1-SM...466322?hash=item464a317692:g:u18AAOSwv9hW31a1

I hope this information is of use and I wish you luck getting into a very enjoyable aspect of the hobby.

Best Regards,
tma

Hello everyone! This may be a long-winded post...

I got bit by the vintage computing bug long ago when I was young due to a mutual interest in anything retro or vintage, and my interest in computers. As such I have a whole rec room filled with VIC20's and C64's, 8088's through Pentium-era systems and parts, and a small horde of era-related software that I haven't played with in forever. Mainly because I spent so much time playing with it years ago.

But I always wanted to go older - an 8080 system - but never got a chance due to the absolute rarity of them. I've never even seen one in real life. But then Grant Stockly came by with his detailed Altair 8800 replica and I realized I might have a chance! But naturally as things go with me, I never had the cash at the time to purchase one. I've been desperately holding out for the chance that he would start producing them again since '07 but doesn't seem like it will happen.

Now there's no way I'm going to drop however many thousands of dollars for a genuine Altair 8800 that pops up every now and then on eBay. Even my strong desire to play with one has limits, but then I recently started thinking, "why not build my own"?

I stumbled across the plentiful resource of S100Computers.com and only recently realized that there is a small community building new cards for S-100 systems, meaning there actually is a way to build one from scratch (figuratively speaking) and I didn't have to wait until my hair turned grey to own an 8080-based system!

But as much as I want to build one, I have to admit that I have never done anything like this before. I've never laid hands on hardware like this, I've never debugged physical cards or circuits, and I've never even soldered electronics other than battery cables for my pickup truck. I do have some experience in building simple circuits in robotics class and assembly programming way back in college ten years ago, but haven't retained much of what I had learned.

I realize that taking on a project like this will involve an intensive study not only in soldering and wiring, but also the intricacies of the associated circuit designs, programming, and oh-so-much more. My end goal is to eventually learn how to design and fabricate my own cards, and potentially even my own software.

So I'd like to ask the more experienced members of this forum, is this a feasible project for someone with my limited experience? How many of you have built your own S-100 system? How difficult (in hindsight) was it to get it operational? Would you have rather bought an existing system or still build your own?

 

[glitch]

As above. Early S-100 boxes are often hacks in the original form, then maybe hacked on by one or more owners in the course of their existence. Even boards all from the same manufacturer may not work together. It's old stuff, so unless you're buying a guaranteed working system and *not modifying it* you are likely to have to make changes and/or repairs.

My first S-100 system was pieced together from stuff I found while working for a surplus electronics dealer, in 2006ish. I'd heard of S-100 systems and knew enough to identify the boards when we came across a pile of them, but had never worked on a S-100 system. It was a Cromemco Z-2D chassis with a Morrow backplane, Cromemco ZPU, and Ithaca Audio dynamic RAM board. As luck would have it, the owner of the building had designed a product that used S-100 boards (though not a S-100 compatible bus) in the 1980's and had spare prototyping boards. I ended up working for him that fall after the surplus electronics guy disappeared and never paid me (he stopped paying his rent, too). I'd hand-built an 8085 based system before this endeavor, and had been an electronics hobbyist for many (10+) years beforehand, plus I was working on a CS degree and working part-time as an analog electronics technician and assembler. The Cromemco system was thoroughly broken, and even with experience, help from someone who had been through that era, and an analog shop full of test gear, it still took months of spare time to get it to actually boot a Cromemco 4FDC I'd purchased separately.

In addition to understanding the basics of electronics, you'll also need a solid understanding of digital logic and the tools and techniques of troubleshooting it. It's certainly not impossible to pick up on your own but there's a lot to be learned. You may be better off starting with a smaller system -- something like an 8080 or 8085 trainer -- and working your way up to bigger systems. Many of the trainers out there, especially vintage ones, have well-written manuals with schematics and program listings, and often include or can be used with an introductory lesson on digital logic and microprocessors. Some come with a lab book that provides exercises and quiz questions.

W.R.T. tools, I use a Hexacon Select-o-Track temperature-controlled soldering station, mostly because that's what we used at the aforementioned analog electronics shop, and we had around 80 pounds of spare tips, so tips were free to me. You can do a lot with a pencil type soldering iron, but make sure it has a grounded tip (3-prong cord). A solderless breadboard is also a very handy tool for learning, especially with older 7400 series logic in DIP packages. You'll need a good multimeter, analog or digital, but nothing extremely fancy -- I like my old Simpson 260 (and its smaller cousin, the 160) as a bulletproof analog meter. You'll also want a logic probe. An oscilloscope can be handy, but isn't strictly necessary, especially when starting out.

 

[Tor]

I've been desperately holding out for the chance that he would start producing them again since '07 but doesn't seem like it will happen.

He did a new run a few years back (mid 2012).

I'm interested in S100 as well, the starting hurdle is a cabinet with an S100 backplane.. somewhere to stick all those nice boards we can build. Not easy to find, it seems (anything off ebay has diastrous shipping costs for non-US locations)

 

[MicrocomputerSolutions]

He did a new run a few years back (mid 2012).

I'm interested in S100 as well, the starting hurdle is a cabinet with an S100 backplane.. somewhere to stick all those nice boards we can build. Not easy to find, it seems (anything off ebay has diastrous shipping costs for non-US locations)

A good/sturdy S-100 Mainframe with motherboard and power supply (not including any S-100 boards or cables) can have a shipping weight of between 50-60lbs. Before I stopped accepting bids from Buyers in Foreign Countries and Shipping to Buyers in Foreign Countries, I had numerous eBay auctions for S-100 Mainframes and Disk Enclosures won by Buyers in Europe and the Far East. Not one of those transactions was ever completed, because all of the Buyers refused to complete the auctions by sending the payment. The shipping charge for a package that weighs 50-60lbs was in the $600 range for the slowest, least expensive methods that I could find. A complete Compupro S-100 System with a Disk Drive Enclosure and disk drives with documentation weighs in the neighborhood of 150lbs and comes in three or four boxes, not including display terminals, printers, or other accessories. Shipping by surface was estimated to cost at least $1200-$1500 (not including any import taxes or duties) for Europe back in the late 1980s.

Outside of eBay I did have a number of customers in Europe and in the Far East who were buying Compupro Components. These customers (in contrast to eBay Buyers) had no problem/complaints completing transactions (mostly for S-100 boards and software), and occasionally sending boards back to me for out-of warranty repairs. These customers never complained about the cost of shipping, but never bought S-100 Mainframes, Disk Enclosures or Complete Systems (they were all maintaining or upgrading their Systems).

If you are outside of the US, and trying to control shipping costs, you might consider having S-100 Mainframes and Disk Enclosures shipped without the original linear power supplies, the shipping weight would drop to 20-30lbs). The power supplies make up the bulk of the weight, and you could buy switching power supplies (locally) to replace the heavy linear power supplies that Compupro (and other S-100 Manufacturers used). Besides, many of these power supplies are 35-40 years old, and the power caps have all tried out. These now, failed/defective power supplies (I noticed many Cromemco Z-2 Mainframes with bad power supply caps as early as 1990). So far, I have not seen a single Compupro Mainframe power supply with defective filter caps. Bad/defective old filter caps in the power supply will damage/destroy S-100 boards by letting the wrong or unfiltered AC onto the DC Bus on the motherboard. Buying new switching power supplies to replace old, malfunctioning linears would be a good way to upgrade the system if you're into switchers.

What you could do is fashion a heavy mounting plate and drill holes that match the bolt pattern in the Mainframe or Disk Enclosure of the original linear power supplies, and drill holes in the mounting plate for the mounting screws of the replacement switching power supply/supplies. That why you can restore the Mainframe to the original configuration without making permanent modifications to accommodate the replacement power supply/supplies.

 

[ArchieX]

Wow. Thanks for all the replies and information so far, everyone! That certainly is alot of information to digest (as expected) but I do appreciate the personal experiences shared - that's exactly what I wanted!

I of course was anticipating building a system to be a very challenging experience. I do enjoy diving into something head-first, especially when I have such little hands-on experience because in my opinion that's the best way to learn.

And that's also the reason why I was holding out specifically for an Altair 8800 because of the sheer plethora of technical manuals and guides available for it, in combination of it being a seemingly-simplified design layout when compared to later S-100 systems I've seen.

As for buying an existing S-100 system, would I be incorrect in interpreting some of the responses regarding that option to be along the lines of "if I buy one, I also buy the problems the previous owner had with it"? Which makes sense, but also perfectly describes my house, my truck and my snowmobile, hehe! Hardly new ground for me, but again, would make an already steep learning curve even steeper than if I were to build one up and learn from my own mistakes, rather than someone elses.

Certainly much to think about. But worry not, for this is not a project I would stampede into without doing much more thinking.

Thanks for the replies so far!

 

[glitch]

in combination of it being a seemingly-simplified design layout when compared to later S-100 systems I've seen

Yes and no. They got to make a lot of the early mistakes. When you really dig into a lot of MITS' designs, it almost seems like some bits of the system were designed by people who had no idea what they were doing. Mostly I fix and sell Altair parts and systems, as I prefer basically everyone else's hardware!

 

[MicrocomputerSolutions]

Are you interested in the Journey or the Destination?

If you are more interested in the Journey, you might be saying that you want to jump in with a pile of old, non-working junk, learning how S-100, and the individual board are supposed to work, fixing the boards, then learning software so you can make the thing work. You like struggling to learn something new, and you like/enjoy pounding your head against the walls for hours at a time figuring how something works, and why it doesn't. You don't know what you're going to do when you get there, but you're enjoying the trip.

If the destination interests you more, then you're about gathering working parts, assembling the thing, and operating it. Pounding your head against the walls for hours at a time, trying to understand how something is supposed to work, and figuring out why it doesn't is NOT WHAT YOU WANT TO DO. You want the thing to work, and you want to learn to work it.

In the first case, you probably want to buy a pile of boards, an old S-100 mainframe, figure out why they don't work, and fix them, using a bench covered with test equipment you bought or have been itching to buy so you can take-on this sort of project.

In the second case, you should probably figure out what you will need to operate the computer, then buy working boards or a whole system that will allow you to run the sort of programs you envision yourself running. You don't want to work on a pile of scrap, you want to operate.

 

[MicrocomputerSolutions]

There is a complete S-100 Mainframe (No Name Computer) with a CCS (california computer systems) board set and two integrated Shugart 8" single-sided double density floppy drives that is for sale on eBay. The auction just ended for about the 2nd/3rd time with no bidders. The Seller has been gradually lowering his asking price.

The No Name Computer Mainframe is a solid design made out of a heavy gauge metal/aluminum. CCS was a major S-100 Player for a short period of time (early 80s) before imploding. The quality of boards was a little thin, but no different from many other brands of S-100 boards of the time (gold plated edge connectors wear out rather easily). I've been using CCS Terminator/Extender Boards since they were introduced, and the boards oxidize excessively/badly IMHO (even in an office environment), but I'm used to Compupro boards which are very high quality.

The Seller notes that he/she does not think the computer is working, and their isn't any software/operating system included. If you are interested in buying that computer from the Seller, I do have an Original CCS CPM Distribution Disk that I would consider selling so you would/have everything you need to get the computer up after you fix the board/s and clean/service the drives

The single-sided 8" floppy drives would probably be better if they were cleaned and serviced before being put back into service. I can tell that one of the drives is VERY OLD from the appearance (mid-70s?). manufacture?).

Last asking price was $475 with no takers. I interested, you might consider messaging the Seller to see if he would accept a Buy It Now Offer of $350-$400 or so (I'd start at $350, and expect a Counter-Offer), and maybe you can get it without going through the auction process.

 

[ArchieX]

Once again, thanks for the advice and responses so far, however I think that some of you are offering a bit too much beyond my initial - and granted, very simplistic - ideas in regards to my first venture into S-100 systems.

Firstly for those of you that have offered me items for sale, understand that I am not in any position to make any large purchases any time soon. Otherwise I'd be staring at that Altair I've always wanted by now, hehe. And plus in regards to previously mentioned shipping charges, I'm located 12 hours north of Toronto, which would give indication as to how much shipping a large existing S-100 chassis would be.

And secondly, perhaps I should have elaborated more on my ideas for building my own. S100Computers.com states it offers individual cards for low cost (when stock in available during group purchases), usually in the vicinity of $20 to $30. I would start by purchasing the 9-slot backpane and building my own chassis around it, then wiring in my power supply. Then progress with one card at a time, such as the front panel controller, the 8080 CPU card, the memory, and so on until - with much education, persistence and luck - I might be able to get a functioning system that I could experiment more on while keeping the costs down to an acceptable minimum.

And in regards to my lack of experience and skills, such as soldering, I ofcourse would not just dive in and start assembling such a project without at first investing much practice into such things. And also with time I would have to slowly invest into the tools necessary to help furnish such a project.

I was in search of advice as to what avenues I should explore but am not interested in purchasing any equipment at this time. Once I decide on which course I want to take (I might even decide not to build one after all), I will in turn do the associated research before committing myself to any ordering or purchases.

Thanks for the replies, everyone. I already have alot to digest with what's just been posted here, hehe.

 

[MicrocomputerSolutions]

If you want to try doing it the hard way, here's an inexpensive power supply to get you started. This is way less expensive than you could build one yourself for.

http://www.ebay.com/itm/121852143882?_trksid=p2060353.m1438.l2649&ssPageName=STRK:MEBIDX:IT

$32.95, plus shipping, and it will easily power a 10-slot motherboard.

But you should know, that while you can get some stripped new S-100 cards for $40 or less delivered, the parts you are going to need to buy will cost you many times that amount.

By the time you get around to finishing those bare boards I would expect you to have at least $150-$200 (minimum in each board).

 

[glitch]

But you should know, that while you can get some stripped new S-100 cards for $40 or less delivered, the parts you are going to need to buy will cost you many times that amount.

Especially if you don't already have a selection of basic 7400 series TTL to pick from. Count on $40-60 in parts depending on the board, if you're ordering just enough for one board.

 

[candrews]

Build. No question about it. but, personally, i like building things with the 8085 much more than the 8080.

i have a number of vintage s-100 but i use a 'current issue' S-100 system (s100computers.com) for working on boards, new designs, etc., since there is no way that i want to put an unproven board into my altair. learning to solder is not a major task, just get a decent soldering station and practice before moving on to a board you really care about.

i would also recommend getting something like an Intel SDK-85 for learning. even after 35-40 years my SDKs are a great source of entertainment and i use them for testing new designs. SDK-85s can be had for under $100 on ebay now and again. ignore the $250-$2000 ones. dont worry too much about the condition, they are simple to work on, parts are cheap and available, and will be good training.