Anyone use(d) Excel 2007? How does it compare with previous versions? Specifically, how does it perform - improved? It's reckoned to utilise the capabilities of dual processors and hence more efficient. But how is this manifest? Tried the trial version but it didn't work (and caused other problems). Any comments greatly appreciated. Thank you. Grant.
I use Excel extensively. I migrated to Excel 2007 from 2003 and used it for a while. Pros: 1. The files are considerably smaller. So saving times have been considerably improved. 2. Lots of rows available. No more 64K limit. Cons: 1. Did not notice any measurable macro running performance improvements. You need to realise that VB is still single threaded. So all the multithreaded stuff applies only to the Excel formulas, so there may be some (considerable?) performance improvements in the sheet calculations. So what?! 2. Most of my spreadsheets became a bit more unstable and crashed more frequently, particularly when editing the VBA code. 3. Did not manage to get RTD working at all. #N/A# As a result I had no choice but to rollback. I am using however the rest of the office suite i.e. Outlook 2007, Word 2007, etc. JMHO
Trademomentum, Thank you for your reply. âall the multithreaded stuff applies only to the Excel formulas, so there may be some (considerable?) performance improvements in the sheet calculations. So what?!â This is perhaps the crux. Iâm trying to analyse time series data and have to keep reducing the amount of data, otherwise Iâm waiting for calculation. That aside, your reply hardly inspires confidence. Indeed, a review I read recommended upgrading only if the multi-threading aspect was important. I first used Lotus 1-2-3 in 1990 (contained on a single floppy), no hard drive and 64k memory. Given I could do then what I am doing now (more or less), I wonder if the increase in computing power is commensurate, given the increased resource requirements to do the same task? I think thatâs a âNo sale at the moment, Mr Gatesâ. Grant.
I migrated a set of 5 min ES data from the last 6 years over from Excel 2003 to 2007. Before it was spread over multiple sheets and now it fits on one sheet! This is a big plus for me. Formulas on the full data set appear to run faster than before and saving seems faster, but I did not benchmark any of this. I was also able to create a 1600x1600 correlation matrix, which would have been impossible in 2003 because of the columns limitation (it took several hours to run). I have run some old VBA code without any problem so far, but I haven't developed anything new in the 2007 IDE. I had some trouble getting add-ins to load but it may have been a learning curve problem. Bottom Line: If you need to handle large data sets on one sheet, it sure beats writing code that has to iterate over multiple sheets. Otherwise for data analysis 2003 works just fine.
MicroSloth screwed-up (again)....they had a great opportunity to improve Excel...and besides the row/column limit expansion, the only big improvement is the document security in Excel 2007 which is unhackable (so far). It employs a Vista-like encryption which is said to be unbeatable. HOWEVER, they did not provide the same security for VBA code when XLA "add-ins" are used. This was a huge mistake IMHO. VBA code in XLS files is secure however. Strange. But that word fits Microsoft to a tee.
This may be a very obvious point to make but one of the best ways of speeding things up is to do the calcs inside a VBA routine rather than on the sheet. You can copy a whole range of data into an array in one single instruction. Then you can loop down the data doing whatever. Then it is a single line of code to copy the stuff back. The problem with doing lots of calcs on a sheet is that when a cell changes, it goes off to recalc a lot of other cells (in auto calc mode at any rate). And it goes through this every time a cell changes. Not very efficient. It seems that the sheet updating is the slowest part of the operation. That is why working with the data in an array is a lot quicker - no screen updating. I know this is not very helpful if you do not know how to write VBA, but seemed a relevant point to make in response to what you are trying to do in Excel.
Swandro, I think youâre absolutely right. My VBA is basic and I keep delaying investing in a decent book (John Walkenbach's Excel 2003 Power Programming with VBA seems highly regarded â the 2007 version is only a few pages difference). Iâve done the calcs now but will need to address some of the parameters in a modified form for running in real-time. In the meantime, Iâve printed your reply for reference. Thanks for the suggestion. Grant.
Using Excel 2007 Trial Download (beta and Refresh versions were / are buggy) with dual core and with 2GB of memory... Using RTD module addin that works fine with it... pulling in 1600 stocks using these fields: change, last, volume, etc to build custom sp500, sp400, sp600 Advance Decline, VolumeDelta, NetChange, Fast Cash, Trin, etc. charts This way i am not hostage to the bullsh**t disinformation of watching the NYSE ADD, VolD and other indicators... that are measuring 3000 stocks and not only the index that i might be trading... Set the multi core option to yes to optimize the use of my dual core... not sure how to measure that but seems to be a bit better than the setup in 2003... Do not like the ribbon... hope it dies a quick death but not betting on it... Using VBA code to pull calculated column data from the 07 Excel sheet and throw it into Sierra Charts every 2 seconds... works great... other wise it is tiresome and lacking to properly chart multiple data streams inside of Excel and better to do it with a outside charting program but the only one i have found that works perfect is Sierra Charts in a disconnected state... although you could problably have it pulling data from a data feed and still ALSo throw the Excel data into one or a dozen charts no prob... HTH cj...
Here is an interesting link on some of the improvements in Excel 2007: http://msdn2.microsoft.com/en-us/library/aa730921.aspx, with this section dealing with the multithreaded capabilities http://msdn2.microsoft.com/en-us/li...ice2007excelperf_ExcelPerformanceImprovements.
Edgehunter Using RTD on 1600 symbols is fine? I attended a seminar for hedge fund users of Excel last night at Microsofts Manhattan space in midtown I asked a developer if there was any improvement in the CPU usage with 2007 v 2003. 2003 absolutely crushes your CPU in my experience He confirmed that 2007 takes advantage of the dual processor whereas 2003 did not but he sounded negative as to any real speed improvement. Are you seeing differently? I use 2000 equity symbols with 5 columns of live data and I have a P4 dual processor 3.2 and 4 GB RAM. Thanks for any response