Does ANY Beta Tester like PRIME?

[TedDiehl,PhD]

Is there any BETA Tester out there that thinks this is anywhere close to being a viable product to release?

I cannot imagine any existing Mathcad user switching to this. It is essentially INCOMPATIBLE with existing Mathcad documents (unless you have very simple docs with NO images, no graphs, no text region formatting, limited use of arrays and equations). Check out the attached documents.

I cannot imagine anyone doing anything except the most simplistic of calculations with Prime.

Please - somebody show me the light!
I am Mathcad's oldest user and its #1 fan - heck I wrote a toolbox for it (www.BodieTech.com).

I am frustrated because PTC talks about new features (the ribbon - it makes use slow and too much mousing; mixed arrays in units - great accept when basic stuff does not work or is severely weakened, what is the point, ...)

Below is a brief summary on my 5 hour ordeal with Beta testing.

I just spent 5 terrible hours with Prime 1.0 Beta. It lacks many BASIC features that exist in current Mathcad (let alone "advanced" features like programming, etc. which I knew would be missing). I have NO (that is ZERO) documents that open and work/format reasonably in Prime (that means 0% backward compatibility). The 2-D graphing is very weak (missing many common graphing features we already had in Mathcad and exist other software). Graphing also starts to choke HARD and Hangs FOREVER when plotting data vectors larger than 10,000 points (I often use 1 million in Mathcad 14). You cannot even copy the graph into PPT (most of us need to present results to management!).

Typical experience of mine: I open a very simple Mathcad 14 doc in Prime - I find that Prime supports NONE of the Mathcad 14 text formatting, bolding, underline, etc. The result is terrible and the computational result was essentially incompatible (errors on a document that computed fine). The document would NOT compute properly because Prime does NOT support arrays with units that are raised to powers without using the Vectorize operator (this has worked fine in Mathcad for many many versions). The 2-D graphs do not even show up and the images I had in the original document - gone! The list of problems and gripes just goes on and on.

AM I DEMANDING TOO MUCH?

Ted Diehl
Bodie Technology, Inc.

[TedDiehl,PhD]

More files that I tested

[TedDiehl,PhD]

And some more

[TedDiehl,PhD]

More files - Had to zip it because Prime made the file too big - yet another gripe!

[RichardJ]

We are missing the data file for this one.

Richard

[TedDiehl,PhD]

On 8/23/2009 4:51:54 PM, rijackson wrote:
>We are missing the data file
>for this one.
>
>Richard

Richard, et. al.,

Here are the files I was talking about, all zipped up this time

[SteveKeen]

I have to agree Ted. I have approached every new beta of Mathcad with enthusiasm; that enthusiasm has rapidly evaporated this time as I explored Prime.

I think there are improvements in text handling, but they are relatively minor compared to the many features of Mathcad 14 and below that are dysfunctional in Prime 1.0.

I expected substantial improvement in the interface for graphs, but apart from the ability to relocate axes, there seems to be precious little added to graphing and a hell of a lot lost.

I can't even say whether there have been improvements in data importing because my attempts to import a substantial data file (Dow Jones Historical data) have crashed Prime.

For my money, this is a "back to the drawing board" product.

[PhilipLeitch]

Just remember that every project seems like a failure in the middle. If every project was abandond and started over just because it "looked" like a failure, nothing would ever get done.

I'm going to put my hand up to the title question, "Does ANY Beta Tester like PRIME?"

Yes - I do. I like it for the new interface and for the potential it introduces.

But as for the question you first ask in your post:
"Is there any BETA Tester out there that thinks this is anywhere close to being a viable product to release?"

I'll put my hand down for that. All your points are valid.

What's more, we are all saying how essential specific things are from Mathcad 14 - things that have been in the software for more than a decade) - and we can see that they are to be included in future versions, but we have no idea about that time frame. Is Prime 2.0 due in 6 months, 12 months, 24 months or more???

Also, maybe Prime 1.0 is an initial version that is designed as an add-on to their other products but isn't considered a replacement for Mathcad 14 for another 12-24 months.

I hope that's the case... otherwise it just doesn't make sense. I can't use Prime for my work because of what it is missing.

Philip
___________________
Nobody can hear you scream in Euclidean space.

[TedDiehl,PhD]

I agree with Phillip that a new easier-to-use interface is needed and Prime's equation editor appears to be an improvement (if you ignore the constant zooming required). At 1st look I thought the 2-D graphs were an improvement, UNTIL I tried to do much with them and found out all the stuff that was lacking and their inablity to work with any sizable datasets.

Bottom line - with all the stuff that is missing in Prime, and I am not even talking about the "advanced stuff" like programming, etc. - Prime as we see it know will be of virtually no interest to any existing Mathcad user's. I cannot believe people would pay $$ for Prime and port to it from Mathcad. I cannot even see how new engineers would be attracted to it. Perhaps some Pro/E user's that desire to drive a Pro/E geometry with a simple stress formula. Having said that, I know many other engineers using Pro/E that would find Prime too crippled compared to their current versions of Mathcad.

If someone thinks otherwise, please speak up. I would really like to see the bright spot, honestly.

Also, if you agree with me that Prime is not anywhere near ready for release, please speak up and be counted.

It is important that PTC Development hears your views clearly on this, whichever way you view it.

Ted Diehl

[StuartBruff]

On 8/23/2009 9:26:46 PM, diehlted wrote:
== I cannot even see how new engineers would be
attracted to it.

As mentioned on another thread, the target
audience might not be as 'conceptually advanced'
as one might believe. For every McEnroe-like call
of "you cannot be serious", there might be a few
sighs of relief. For example, the lack of global
variables is deliberate as they were confusing to
many people.

In addition, Prime contains an awful lot of
functions that were only available through
extension packs. This reduces the pressure to have
programming as some fairly interesting stuff can
be done directly.

I note the opening words on the Spirit help:

Use Mathcad to perform, document, and share
your calculations and engineering design work.
Mathcad integrates engineering-oriented
mathematical notation, full-featured word
processing, and 2D plots and images in a single
worksheet, making it easy to visualize, verify,
and document your work then share it with others.

I look forward to the automatic
paragraph/figure/table numbering that 'full-
featured' word processors provide and engineering-
oriented mathematical notation, such as
J_ν(x) and tensors.

Stuart

[RichardJ]

On 8/24/2009 2:56:25 AM, stuartafbruff wrote:
> For example, the lack
>of global
>variables is deliberate as they were
>confusing to
>many people.

Yes, and that is a very slippery slope indeed! Once the decision is made to remove useful features because they might be confusing to a novice user, where does it end? If a user doesn't understand a particular feature they will likely not use it. If a more advanced user includes that feature in a worksheet then they presumably did so for a reason, and it is not realistic to stipulate that an advanced user can only use features that a novice can then understand. If that's going to be the criterion for deciding what features to remove, they had better remove range variables too, which confuse a heck of a lot more people than global equals does.

Removing global equals will do little to help new users, and will remove something that is useful to advanced users. It will also break a vast number of existing worksheets. Hands up everyone that want's to go through all their existing worksheets and retype all the global equals. And all the math in text regions (something else that I believe was deliberately removed). Those two things alone would break most of my existing worksheets. Of the many other missing basic features (such as column operators on the lhs of an assignment) I can only assume at least some of them have been removed by choice, and they will add to the casualty list.

Richard

[IRstuff]

I agree with the need for global variables. One of the biggest issues has always been the linear flow of the worksheets. This means that at the end of a long sheet, you no longer have access to the input parameters, but you could have used globals at the bottom of the sheet, so that you can see their effects on the final result.

Excel allows you to do something similar, since you can reference the final result at the beginning of the sheet, or place input parameters wherever you want on the sheet.


TTFN,
Eden

[StuartBruff]

On 8/23/2009 8:32:16 PM, pleitch wrote:
>>Just remember that every project seems like a
failure in the middle. If every project was
abandond and started over just because it "looked"
like a failure, nothing would ever get done.

Yes. Or 'Maintenance of Aim' is a good principle
to apply outside of war making.

(of course, this should not be confused with
'Maintenance of Stupidity', although there can be
a very find dividing line between the two)

>>I'm going to put my hand up to the title
question, "Does ANY Beta Tester like PRIME?"
>>Yes - I do. I like it for the new interface and
for the potential it introduces.

See http://collab.mathsoft.com/read?127458,103 for
some positive SUC removal examples.

Plus mixed quantity arrays

Plus row/column insert/delete operator edit
capability (that was long on the Mathcad wish
list). However, I'd like to have seen row/column
insert/delete functions as well.

The single input sequence for the summation and
product operators is neat, although I am
disappointed that there is no vector product
operator (again, long on my wish list).

On the lines of the sum/product operators, it
would have been nice to have a subvector function
as well as the submatrix function, or a smart
submatrix function that assumed the column range
for a vector.

>But as for the question you first ask in your
post:
>"Is there any BETA Tester out there that thinks
this is anywhere close to being a viable product
to release?"

In its own little way, provided they sort some
issues out, Spirit 1.0 has the potential to be a
viable product. It's just nowhere near ready to
supplant Mathcad.

Stuart

[TedDiehl,PhD]

Stuart said

"
In its own little way, provided they sort some
issues out, Spirit 1.0 has the potential to be a
viable product. It's just nowhere near ready to
supplant Mathcad."

The problem is, there has been NO Mathcad upgrade since 14 (over 2 years ago). How many people are willing to stay with Mathcad if Prime is nowhere near ready to supplant Mathcad and Mathcad is not being enhanced, only maintained?

Also, While I agree it is nice that the extension packs are now native to Prime, DSP tools are pretty weak when you cannot work with data signals containing over 10,000 rows without the worksheet becoming sluggish and the 2-D graphs impossible (especially over 100,000 points). This type of stuff needs to be addressed and verified BEFORE they release Prime.

Lastly, someone mentioned VISTA. I FULLY agree. VISTA got a BAD RAP and I know of no engineers(yes, I am sure there are people) that have installed Vista of their own free will. If Prime get's a Bad rap, it will be very difficult to recover from it.

Ted

[RichardJ]

On 8/23/2009 8:32:16 PM, pleitch wrote:

>Yes - I do. I like it for the
>new interface and for the
>potential it introduces.

What potential is that. I am not saying there isn't any, but it does need to be qualified.


>What's more, we are all saying
>how essential specific things
>are from Mathcad 14 - things
>that have been in the software
>for more than a decade) - and
>we can see that they are to be
>included in future versions,
>but we have no idea about that
>time frame. Is Prime 2.0 due
>in 6 months, 12 months, 24
>months or more???

In many cases we have no idea if they will ever be included. There is a list of things that are planned, but there a lot of missing features that are not on that list. Perhaps the biggest is that it cannot read .mcd or .xmcdz files, and cannot write to any previous format. Since they already have the code written to read and write such files why did they not include this capability? And what about referenced worksheets? Are these going to be permanent deficiencies? These would be in addition to the known backward compatibility bombs of global equals and math in text regions, as well as any future incompatibilities in things like programming.

>Also, maybe Prime 1.0 is an
>initial version that is
>designed as an add-on to their
>other products but isn't
>considered a replacement for
>Mathcad 14 for another 12-24
>months.

I do not believe Prime has any hope whatsoever of replacing MC14 in 24 months, let alone 12. It's been 2 1/2 years since the release of Mathcad 14. If the slew of basic features that are missing (such as global equals, column operators on the lhs of an assignment, etc etc) are missing by choice, and will never be implemented, then we have a gross backwards compatibility problem. If they are missing because they didn't have time in the last 2+ years to implement them, then how long will it take to get larger chunks of work completed, such as decent 2D plot, 3D plots, programming, user DLLs, components, animation, etc etc?

I think we should separate Ted's question into two questions:

1) Where is Prime now

2) Where is it going.

For (1), I would say almost nowhere. No existing Mathcad user will switch to it. I can't imagine any large engineering group would choose this as a new platform for calculations, even if they currently don't use Mathcad, because it's not versatile enough. Not many individual users would want it, when there is free software available that is arguably more capable. Maybe some Pro/E or Windchill users will use it because of the integration?

For (2), I think it's an open question. We can all speculate, but we don't know. Even those that have inside information from PTC don't really know, because they may not have all the information, and PTC can change direction any time they wish. Does it have some potential? Yes. I also like the new editor, and some of the matrix entry procedures that Stuart has pointed out. But I hate the ribbon interface, which turns every operation into 3 mouse clicks: select the tab, select the menu, select the item. What's wrong with one click, as we had before? Or just a keyboard shortcut. Try entering infinity. Or %. Or add a trace to a graph. The lack of certain basic features (especially the ones I know were eliminated deliberately) concerns me a great deal. The timeframe to get to the level of version 14 concerns me even more. So far, I have found zero of my worksheets it can read and produce a working document. Not even the most trivial ones, because there's always some basic feature it falls flat with. Like it contains a graph.

So, a very key question for me is will any version of Prime ever be able to read more than a few percent of the most trivial existing worksheets? If yes, how many years, realistically, is that going to take? I cannot believe it will only be 1-2 years. A related question is how long are existing users prepared to wait to get a viable replacement for MC14? Another year? Another 2 years? Another 4 years?

Richard

[mthackst]

In reply to the question, "Does anyone like Prime?"

Not so far. I don't see anything that's better than any version of Mathcad since 4.0. I've been a heavy user of the product for a long time. There have been changes over the years and generally the adjustments have been managable. Generally, each next product was notably better than the one it replaced.

I honestly don't see anything in this intial beta version that is an improvement over 14 other than the fact that full justification is now available in the text regions.

[I don't really count adding in the Extension Packs -- they could just do that with the existing version.]

Is it viable?

I don't know. At present it's way behind any potential competition.

I will say this. The beta that I'm testing seems less buggy than any previous beta. Usually, there have been things in the first version that just gave wrong answers or fatal errors or other really unpleasant things.

This one just doesn't do as much as I'm used to being able to do, and there are too many times that I'm reaching for the mouse and sifting through a menu to do something that one key-press would have done before.

I hope this is not intended as an almost-final version prior to release.

Mike

[Fred_Kohlhepp]

On 8/24/2009 3:57:46 PM, Michael_T wrote:
>
>I hope this is not intended as
>an almost-final version prior
>to release.
>
>Mike

I agree. If this is the first of a series of beta versions, then there's hope. But based on what's been written in the documentation, and what's been evolving in this forum, Prime will be a child's toy compared with previous versions of Mathcad.

Fred Kohlhepp
fkohlhepp@sikorsky.com

[GuyB]

On 8/24/2009 10:30:10 AM, rijackson wrote:
>....
>I do not believe Prime has any hope
>whatsoever of replacing MC14 in 24
>months, let alone 12. It's been 2 1/2
>years since the release of Mathcad 14.

I am more hopeful, but I could be misguided.

I'm just speculating, of course, but I think all this time has been spent on trying to figure out and implement an 'improved' interface. I think that a lot of the extra functionality _should_ be pretty easy to tack on.

After all, it's not like they've changed the native code as for v12 (brrrr). The code that executes the other things we want can remain the same. Once they smooth out the interface, I'm hoping they can write small modules that communicate outputs from the interface to the existing code inputs without much more effort.

Thus speaks a *non* software engineer,

- Guy

[TedDiehl,PhD]

On 8/26/2009 9:28:44 AM, GuyBeadie wrote:

>I'm just speculating, of course, but I
>think all this time has been spent on
>trying to figure out and implement an
>'improved' interface. I think that a
>lot of the extra functionality _should_
>be pretty easy to tack on.
>
>After all, it's not like they've changed
>the native code as for v12 (brrrr). The
>code that executes the other things we
>want can remain the same. Once they
>smooth out the interface, I'm hoping
>they can write small modules that
>communicate outputs from the interface
>to the existing code inputs without much
>more effort.
>
>Thus speaks a *non* software engineer,
>
> - Guy

Guy - as someone who does develop software, I suspect that PTC has changed a lot of the native code in Prime. I think they have built in many incompatiblities into Prime relative to exsting Mathcad. As evidence, the 2-D graphs are so incompatible as to not being able to read an existing Mathcad 2-D graph. If it was just "user interface" as you suggest, they could have read the minimum info on an existing 2-D graph (what is on the X and Y axis) and at least put it on a live Prime 2-D graph. They did NOT, instead creating an image of the old Graph and expecting the user to re-create ALL their 2-D graphs. I could easily generate a very large list of such items.

Anyone care to comment?

[josephbradley26]

On 8/23/2009 1:44:57 PM, diehlted wrote:
>Is there any BETA Tester out
>there that thinks this is
>anywhere close to being a
>viable product to release?
>
>I cannot imagine any existing
>Mathcad user switching to
>this. It is essentially
>INCOMPATIBLE with existing
>Mathcad documents (unless you
>have very simple docs with NO
>images, no graphs, no text
>region formatting, limited use
>of arrays and equations).
>Check out the attached
>documents.
>
>I cannot imagine anyone doing
>anything except the most
>simplistic of calculations
>with Prime.
>
>Please - somebody show me the
>light!
>I am Mathcad's oldest user and
>its #1 fan - heck I wrote a
>toolbox for it
>(www.BodieTech.com).
>
>I am frustrated because PTC
>talks about new features (the
>ribbon - it makes use slow and
>too much mousing; mixed arrays
>in units - great accept when
>basic stuff does not work or
>is severely weakened, what is
>the point, ...)
>
>Below is a brief summary on my
>5 hour ordeal with Beta
>testing.
>
>I just spent 5 terrible hours
>with Prime 1.0 Beta. It lacks
>many BASIC features that exist
>in current Mathcad (let alone
>"advanced" features like
>programming, etc. which I knew
>would be missing). I have NO
>(that is ZERO) documents that
>open and work/format
>reasonably in Prime (that
>means 0% backward
>compatibility). The 2-D
>graphing is very weak (missing
>many common graphing features
>we already had in Mathcad and
>exist other software).
>Graphing also starts to choke
>HARD and Hangs FOREVER when
>plotting data vectors larger
>than 10,000 points (I often
>use 1 million in Mathcad 14).
>You cannot even copy the graph
>into PPT (most of us need to
>present results to
>management!).
>

Ted
I actually withdrew from testing because I had similar experiences to you. I lasted about twice as long in terms of time invested, as I hoped it would get easier to use with time. However, that did not come to pass.

The potential for developing sheets with DOE functionality had been my current interest, but it is going to be easier to use Minitab, as clumsy as that program is.

Joe

[ptc-1209116]

I vote "No", but I'll offer a conspiracy theory.

PTC's big product is Pro-E, one of the major 3d modelers that mechanical guys use. They get a vaguely Mathcad-like product that is now internally hooked to pro-e and their other products, and they give it away with Pro-E. A few people use it, then they start rolling it out with all the stuff we remember but without the weird inconsistencies we're all programmed to handle. They get all the pro e customers to pay extra for it.

Now it becomes a way to dynamically modify drawings - Temperature t rises, so pressure P goes up, and vessel wall r gets thicker, and so on... And all those relations are written in a document that looks like the text book. Mathcad used to have thing called smart sketch around V8 that worked like this for 2 d sketches.

That's all my theory, anyway.

[LarryLederer-di]

On 8/31/2009 5:55:54 PM, arfnotz wrote:

>Mathcad used to have thing
>called smart sketch around V8
>that worked like this for 2 d
>sketches.
>
>That's all my theory, anyway.

You maybe on to something. Smartsketch ver 4 by Intergraph, and VisSim Plus ver. 5 was packaged with Mathcad (either 2001 or 11.) This seemed to be be the future of the product, app's that had interacting links to Mathcad. change the numbers in one and it changed the numbers in the linked Mathcad worksheet and visa-versa. I thought VisSim worked a little better than Smartsketch with Mathcad but I liked being able to make a quick drawings in Smartsketch embedded it in Mathcad. Still both programs caused crashes that diminished the ideas behind the integration.

[PhilipOakley]

These sort of links are important for 'model driven design/engineering' which is the current paradigm in industry.

You will see many of the other companies pushing their capabilities in this area.

MathCAD always had a blind spot for the 'what happens next' stage. It was OK when the designer was expected to provide reports [e.g. for 'strength of bridge'], but is of no use when the next step is to embed the design in a product [e.g. I have proved a calculation flow, now implement it - see the thread on imaging the flourescant beads].

Both MatLab and Maple are pushing their design flow capability, as are many others. Plus industry is wanting that seamless transition.

If I code an image processing algorithm in mathcad, I can see it works, but I can't deliver it. If I do it in matlab I can't always see it is working properly, but I can deliver it (the all important pay-off). [(I believe)The MathCAD whitboard GUI is VERY important here - all the others use a linear command line script interface, hence the need to link to 2d layout tools (LabView, Smartsketch, etc.)]

That still doesn't mean that this 'beta' release is anything other than a quick look at the first alpha!

Philip Oakley

[Fred_Kohlhepp]

On 8/31/2009 5:55:54 PM, arfnotz wrote:
>I vote "No", but I'll offer a
>conspiracy theory.
>
>PTC's big product is Pro-E,
>one of the major 3d modelers
>that mechanical guys use. They
>get a vaguely Mathcad-like
>product that is now internally
>hooked to pro-e and their
>other products, and they give
>it away with Pro-E. A few
>people use it, then they start
>rolling it out with all the
>stuff we remember but without
>the weird inconsistencies
>we're all programmed to
>handle. They get all the pro e
>customers to pay extra for it.
>
>Now it becomes a way to
>dynamically modify drawings -
>Temperature t rises, so
>pressure P goes up, and vessel
>wall r gets thicker, and so
>on... And all those relations
>are written in a document that
>looks like the text book.
>Mathcad used to have thing
>called smart sketch around V8
>that worked like this for 2 d
>sketches.
>
>That's all my theory, anyway.
>

I'm sure you're right; those of us in industry that knew about Pro-E feared this from the beginning. But does that mean they have to emasculate Mathcad to create this interface?

Fred Kohlhepp
fkohlhepp@sikorsky.com