The Power of Yelp: Designing Chips versus Bathrooms

Not for the first time, it’s become obvious that designing a bathroom is very similar to designing a chip. The effort requires a confident sense of the end-product you’re trying to accomplish – and then a great deal of necessary tedium.

Because these days, bathrooms and chips incorporate a huge number of IP blocks and the process of tracking down the final candidates, reworking the design over and over to see how each candidate fits in, then making the final selection, confirming with each IP vendor that the price and availability of their particular block meets your design budget and schedule, then nailing down the finished design, actually ordering and paying for the IP that will go into the design, and crossing your fingers that nothing untowards has happened between the moment you made your final part/IP selection and the moment the vendor’s supposed to pony up the goods.

The process of creating a bathroom or chip design that includes a great deal of IP is really complex. It requires a lot of shopping, validating, purchasing, receiving and integrating IP into the design.

But there’s another hitch: Let’s say you designed a bathroom that was installed in the house 3 years ago, and now you’re adding a bathroom that you want to look and function in a very similar way.

Or let’s say you want to design a chip based on a design you successfully completed 3 years ago, and now you want your new chip to function in a similar way with a few important tweaks and improvements.

Are the IP blocks you used 3 years ago to do the design still available, or have they been discontinued by the vendor or changed beyond recognition?

Has the migration to smaller process nodes over the last 3 years – or maybe even to bigger, less expensive nodes – made that IP inappropriate for your current design?

Worse yet, has the vendor that sold you that IP gone out of business sometime in the last 3 years, and you’re only now discovering that fact – or been acquired by another IP vendor who has no interest in helping you figure you where in their catalog the IP you used 3 years ago now resides.

Or worstest of all: Did you fail to make adequate notes 3 years ago? And now, unbelievably, you can’t figure how where you got some of the IP you used then, who you bought it from, or how it was modified to fit into your design. Are you a victim today of your own lackadaisical attitude 3 years ago towards documentation?

In the case of my bathroom, 3 years ago I procured parts from Grohe, Hansgrohe, Kohler, American Standard, Delta, Ashbury, Baldwin, Moen, Pella, Afina, Toto, Aco, Panasonic, Levitan, and Benjamin Moore. But the granite, glass, hinges, grout, heater vents, insulation, LED strip lighting and shower-niche form used in the bathroom construction were all provided by the general contractor and I have no information as to who manufactured these things or where they were purchased and/or fabricated.

Now I’m designing a new bathroom, but working with a different general. Some of the materials and services for the current project should be purchased from the same vendors as 3 years ago to keep a consistent look and feel in the final outcome, but the information is missing. This is a problem.

It’s also a problem if you’re designing a chip today that is supposed to look and feel somewhat like the chip your team designed 3 years ago.

Unfortunately, two guys from the team 3 years ago are no longer with the company, their notes left with them, and now some of the IP decisions you must make today will be guesstimates. Hopefully, you’ll make the right guesses and will find the IP and associated vendor that worked well before.

Surely it can’t be that you’ll find yourself completely reinventing the wheel, having to redo dozens of decisions that were made 3 years ago to guarantee equivalent success today. Seriously, how much could things change in 3 years?

In a similar vein, hopefully my general contractor will be able to track down the same glass and granite fabricator, the same tile installer, an equivalent grade of grout, the same dry-wall guys to provide continuity on the wall finishes, and the identical heater grates.

Fortunately in the case of the bathroom, the last 3 years have also seen some good changes. Wall-hung toilets were a novelty 3 years ago. The plumber who installed mine had never worked on one before; there’s an annoying shimmying behind the tile and sheet rock every time it’s flushed. The plumber that comes in to work on the new bathroom should now be more experienced with these devices, and I expect better success.

Things might have changed for the better with your chip design as well. Maybe 3 years ago your engineering manager had not yet imposed strict documentation regulations, but now it’s required of everybody.

Also, maybe the IP vendor who won the bid to provide some particular block 3 years ago, that block that proved wonky in the chip and had to be swapped out, has happily gone out of business. Maybe now you’ll be able to argue for going with the more expensive vendor who has the far better track record and superior product.

And, hopefully the guys who have joined your design team in lieu of the two guys who left have greater skills than the people they replaced. Maybe they’ve got a detailed list of dozens of trusted IP vendors that they’ve worked with before, and personal relationships with AEs at those companies, and will be able to zero in quickly and confidently on the exact vendor and block needed to meet your current design specs.

In the case of home construction, now LED lighting has become the thing in interior design. As opposed to 3 years ago – when it fell to a friend who is a skilled amateur electrician to show the professional electrician how to install recessed LED lighting in a closet or cupboard – hopefully here in 2017, the electrician sub will show up with a complete set LED lighting installation skills and will be able to do the job without assistance.

Of course, there are still challenges.

Between Grohe, Hansgrohe, Kohler, American Standard, Delta, Ashbury, Baldwin, Moen, Pella, Afina, Toto, Aco, Panasonic, Levitan, and Benjamin Moore – not all of these companies are actually producing the exact products I selected for the bathroom constructed 3 years ago. Their catalogs have evolved and I have no choice but to evolve with them.

Although it is possible to go online and try to find parts for the new bathroom from resale companies who have back stock from 3 years ago, it’s really better to buy what’s currently available from the manufacturers directly, or through certified reps, because then the products come complete with guarantees and warranties.

Similarly, rather than buy outdated parts, your IP choices for your new design are also better if they come fresh from the vendor with guarantees and warranties that are current here in 2017.

But wait, what’s that? You say your IP blocks don’t come with the kinds of guarantees and warranties that are standard with sinks, medicine cabinets, faucets, plumbing fixtures, door hardware, windows, ventilation fans, and paint?

You must be kidding.

How do you move forward on a chip design, with a budget orders of magnitude bigger than that for my bathroom, without guarantees from the vendors that all of the dozens of IP blocks you bought from them and now are going to insert into your design will actually function as promised when the whole thing is manufactured and put out to the market?

How do you do it?

Well clearly you do, but the effort must entail a lot of long days, endless discussions of risk mitigation, many sleepless nights, and a lot of dumb luck to tape out a design that works according to the specs with so many IP blocks that are not guaranteed by the vendors to work in your particular design.

So – there are similarities between assembling a bathroom out of dozens of pieces of IP procured from a host of different vendors and assembling a chip out of many, many dozens of IP blocks procured from multiple vendors. But clearly the comparison stops there.

I can hold the general contractor, all of the subs, and the IP vendors responsible for the success of my bathroom construction. More importantly, I can hold Yelp over their heads if anything goes wrong.

What can you hold over the heads of your IP vendors when things go wrong? Not much, as far as I can see. They can always claim the fault was in your design, or worse yet – refuse to do business with you going forward if you complain too loudly or too publicly.

As far as I can tell, the IP vendors have very little skin in the game, but you have lots. You have lots to gain if the chip succeeds, and even more to lose – including your job – if some of the IP onboard fails to integrate or perform as promised.

If my bathroom fails, I’ll hound the general until the problems are fixed. If your chip fails, who are you going to hound?

Turns out, there’s really nothing similar between bathroom design and chip design. Nothing at all.

*****************

Tags: chip design, IP integration

This entry was posted on Thursday, March 9th, 2017 at 7:05 pm. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.