b'windows, which was the point, after all. I didntCODE OFFICIALS AND CONSUMER GROUPS SAW THE VALUE see the value in spending upwards of a million dollars just to make it look pretty. IN GETTING ACCURATE WINDOW PERFORMANCE NUMBERS, In fact, from EMRs side, there was someAND MANUFACTURERS WERE HAPPY NOT TO HAVE TO concern about the availability of funding forTEST EVERY WINDOW IN THEIR CATALOGUES.future versions of the software. EMR spon-sored a strategic meeting of all the people we knew who were involved in some aspect of window simulation. People from Berkeley, theUniversity of Waterloo, we even developed aoften confuse precision and accuracy, but University of Waterloo, the University of Massa- method for predicting window condensationtheyre not the same thing. We followed chusetts, some private-sector U.S. concerns,potential, which provided good agreementASTM convention that says the results of a and good old Enermodal, convened in a townwith measurement, whether via thermo- calculation cant imply greater precision than called Elora, just north of Waterloo. We agreedcouple (Figure 4) or infrared (Figure 5). Thisthe inputs used in the calculation. Thermal that public funds were limited, and that publicprocedure was included in the 1994 CSA A440conductivity of materials is only reported to money shouldnt be wasted, and that weStandard as an accepted method to evaluatetwo decimals at bestyou cant have more should all collaborate and not waste time tryingwindow condensation potential, but was votedprecision than those inputs.to duplicate each others efforts. Everyoneout later that year, largely by manufacturers.shook hands all aroundand the next thingEventually, it became clear which way the we heard out of Berkeley was THERM, a directEventually, though, big money won. The U.S.wind was blowing, and EMR wasnt interested copy of FRAME (except with prettier pictures),government heavily promoted THERM, andin funding any more software development. and pretty much the opposite of what we allEMR apparently didnt want to outspend theThe FRAME user base gradually shrank, and agreed to do. U.S. Department of Energy. So THERM gradu- THERM became the standard.However, FRAME was still the more popularally took market share. Still, the U.S.-based NFRC would not have choice. It was still fast, accurate and easy to use,THERM reported results to four decimal places,existed were it not for a very successful (but and it had over 90 per cent of the market shareso people took that to mean it was morecrude!) little Canuck programand Ill bet for evaluating window thermal performance.accurate than FRAME, which only carried twomost people reading this piece had no idea it Working with Professor John Wright at thedecimal places. Thats understandablepeopleeven existed. FALL/WINTER 2019 13'