Mar (03-05) - For want of a nail

March, 2005 Editorial

I think we are (over)due for a big run on IBM 6400's.

The supply of printbands for IBM 3262, 5262, 4234, 4245, 6252, 6262 and other Band Printers has all but evaporated. We've had several on order since last July! This is beginning to come to a head as users discover too late that their "old iron" is useless in the absence of this consumable. The process of making a printband is not unlike the manufacturing of a computer IC Chip. There is a mask applied to the raw material and successive acid baths remove the unwanted material. This process "burns" a printband from a thicker piece of metal. The process is messy, time consuming, and dangerous. The environmental concerns are striking. The firm that has the infrastructure to make printbands has to consider the implications and costs of rebuilding their hurricane-destroyed plant, relative to the demand in an aging installed base. So what do we recommend? The sweet spot in the market is the IBM 6400-015. The 6400-010 is a sweet machine also, but the supplies are quite tight of that model.

Be warned: If you intend to seize the opportunity to replace your crippled band printer with a device intended to run ethernet, be careful what you wish for! Any attempt to print from an AS/400 or Z series/ s390 mainframe will have a problem. I have written and "taught" extensively over the past several years about the Black Hole of print. If you intend to print from either of the above mentioned sources, please read my white paper:

Moving From Coax (3270) or Twinax (5250) SNA to TCP/IP Ethernet for Host Connectivity: A Strategic Guide to Implementation
and/or call to discuss the options. Don't make an expensive and embarrassing mistake. Get the info you need to make the right choice.

And here's another "curve ball": There aren't any channel-attached impact printers left to offer!

4247's are getting to be like hen's teeth. The few we have found we have inventoried, but its getting "skinny" out here. Use IBM 4230-102 or 4i3 for twinax IPDS, 4230-202 or 5i3 for coax IPDS, or ethernet IPDS with out attachment card. Otherwise, you'll probably try to make an ASCII impact printer work on your mainframe.... good luck.

Seems that another well-respected dealer is getting out of the printer market. No one ever said that this is an easy market!

Last month's article made mention of the nervousness of the PC market due to IBM's Lenovo sale. I received a comment from a colleague and want to comment on another perspective. I'll give you a synopsis:

The market for used PC's has gotten much firmer over the past several years. There are several reasons which should be noted for this market shift:

1) Changes to the market have resulted in a smaller number of dealers and leasing companies. There are just fewer players. The survivors are those with patience and conservative expectations. The go-go days of seemingly infinite supplies have dwindled and the remaining players are less reckless in their deals.

2) The big names haven't changed in the past 15 years. The big three have always been there. Even IBM's sale to Lenovo won't show a marked change to IBM's distant third place. Thiner production and manufacturing had been done in China and Asia anyway. Dell is the 800 pound gorilla. The quality and stability projected by Dell has created a surprisingly orderly and predictable market.

3) The turnover rate has slowed. Users are finding that the marginal improvements in processing power and storage have less value relative to the cost of an incremental upgrade. Gear is not being "churned" at lease expiration, resulting in less equipment coming back to market. The value received from current generation of equipment seems to outweigh immediate upgrades. The plateau of technology has contributed to the change in market performance. The new products continue to push down prices for second-market equipment, but a slower rate than in prior years. As the new multiprocessor core units hit market in the future, their price points may cause a readjustment but overall trend should be consistent.

4) Disintermediation in the industry has taken many levels of brokers and dealers out of the loop, reducing the transaction costs borne by the final end user.

5) Volumes are stable. Manufacturer (Vendor) lease programs have eaten away at the availability of second market products hitting US markets. This will dry up supplies over the long run, but it's such a big market that it had been hard to notice over time. This attrition from vendor leases has been successful in removing the vendor's cannibalization risk.

Certainly there are more reasons, but these seem to be the primary justification for this newly tight (?) and stable PC market.