I HAVE TRASHED MULTICHARTS

Questions about MultiCharts and user contributed studies.
khalaad
Posts: 323
Joined: 07 Jan 2007
Location: Lahore, Pakistan
Has thanked: 64 times
Been thanked: 57 times

I HAVE TRASHED MULTICHARTS

Postby khalaad » 22 Aug 2008

Marina,

You run a marathon (of sincere) effort every day.

I honestly think you win it every day.

This is why I am harsh: I judge MultiCharts on the basis of claims it makes about its features, and the understanding it offers the users regarding its future development.

In my opinion on both the above counts MultiCharts needs to put its act together. I have thought about writing detailed 'what's wrong'; then given up the idea because I do not believe it is going to change anything.

Yes, I paid only $400 for my MultiCharts; but I paid it at a time when MultiCharts couldn't even correctly draw a simple barchart.

And since the beginning of 2007 I am waiting for certain perennially imminent developments to happen; I do not want to still be waiting well into 2009.

Had I not put in lot of time and effort into EasyLanguage, I would have trashed MultiCharts long ago. It was only today I became reconciled to such a loss and I am very relieved at DOING A COMPLETE MULTICHARTS CLEANUP from my computer.

I have looked at NT, and two days ago I started checking out SmartQuant. Even if both are no better they are at least both less expensive than MultiCharts!

Time to learn C#.

Khalid

User avatar
TJ
Posts: 7740
Joined: 29 Aug 2006
Location: Global Citizen
Has thanked: 1033 times
Been thanked: 2221 times

Postby TJ » 22 Aug 2008

I share your pain.

It is frustrating at times.

Some of the simple things frustrate me most.
e.g the manual scale that does not scale manually...

sigh...

Tresor
Posts: 1104
Joined: 29 Mar 2008
Has thanked: 12 times
Been thanked: 53 times

Postby Tresor » 22 Aug 2008

Khalid,

What happened?

fs

Postby fs » 22 Aug 2008

Khalid

If you search my posts, you will see that I have reached the same conclusion a few weeks ago. The only thing I use MC now for is basic real time charting. MC used to have a lot promise and this also why I got in early. Unfortunately, the status of MC years later is still that it have a lot of promise. They implement features just for the sake of listing it on the website to claim they have a certain feature. If it has bugs, too bad... they are working on the next feature which have "higher priority". In the end you sit with a product with lots of half working features.

Tresor, if you want to use it for backtesting. Be careful. It is simply not accurate. I have spent weeks e-mailing Marina with backtesting results, showing her that they calculate drawdowns incorrectly and she kept on insisting that it is correct. Only after I posted on Elitetrader, did Andrew confirm that they do not calculate it correctly. Who know when and if this will be fixed. But I don't care anymore. I am using other platforms now for backtesting.

This is the frustration with MC/TS Support. They rather want to implement bells and whistles to make it look pretty instead of focusing on doing the basic things right.

Tresor
Posts: 1104
Joined: 29 Mar 2008
Has thanked: 12 times
Been thanked: 53 times

Postby Tresor » 22 Aug 2008

Tresor, if you want to use it for backtesting. Be careful. It is simply not accurate. I have spent weeks e-mailing Marina with backtesting results, showing her that they calculate drawdowns incorrectly and she kept on insisting that it is correct. Only after I posted on Elitetrader, did Andrew confirm that they do not calculate it correctly. Who know when and if this will be fixed. But I don't care anymore. I am using other platforms now for backtesting.
I remember your posts on drawdowns calculation on elitetrader.com. This issue also makes my ass aching. The other issue is the logic that the Backtester uses when reversing a position. The Backtester thinks there is not sufficient cash for a reversal and one has to artificially double one's account to perform reversals.

What is the ''margin'' dialogue box in QuoteManager for, if it is not used in bactesting / optimization?

I have not used the Backtester for equities. Only for futures contracts. The Backtester makes it practically impossible to backtest futures properly. I do agree with you on bells and whistles. Why not focus the Bcktester first (and finally fix it) and then take care other feature requests?

Regards

khalaad
Posts: 323
Joined: 07 Jan 2007
Location: Lahore, Pakistan
Has thanked: 64 times
Been thanked: 57 times

THANK YOU GUYS

Postby khalaad » 22 Aug 2008

Hello!

Thank you guys for upholding me.

MultiCharts has become SMUG; the increasing sales have gone to their head and they are not able to handle even this small degree of recent success.

There was a time the MultiCharts support was beyond-the-call-of-duty excellent. Now MultiCharts cross examines before it even would consider investegating a very real and long running problem.

Even if one proves MultiCharts guilty beyond any reasonable doubt they don't admit a problem exists. The matter of drawdown calculations is a present and increasingly known example.

MultiCharts is now primarily focused on SHOW features -- these require greater development time -- not features which (take less development time) and would facilitate a user trade with greater efficency, increased comfort and allow him/her to generate higher profits

I want ANYTHING which helps me trade:

more efficiently (if one does not know ones true drawdowns, how does one properly position size?);
without stress (it is distressing when manual scale that does not scale manually);
without resorting to an increasingly uncaring and prosecutorial support ("elaborate" what Andrew has already admitted as a problem on another forum!).

Khalid

User avatar
Marina Pashkova
Posts: 2758
Joined: 27 Jul 2007

Postby Marina Pashkova » 25 Aug 2008

Hi everybody,

I just decided to answer you all guys in the same post.

Fanus, we have admitted that in the case that you are referring to we were wrong: the basic specifications for the calculation of drawdown were in principle different from what you were saying. All my and our programmers' responses were based on those specifications. You on the one hand and our team on the other were pretty much talking about different things. It did take us a while to understand that we were wrong but once we realized what exactly was wrong, we acknowledged the problem. Moreover, this problem has already been fixed in 4.0 beta 1. Once again, I can only apologize for the confusion and misunderstanding.

Tresor,

I am not sure how the above problem that fanus helped us identify is related to reversing the positions in backtesting. Anyway, the fact that the logic of reversing the positions is wrong, is a known issue and will be fixed in one of the future versions.

The margin dialogue box has been made to be used in future implementations of MultiCharts.

We did admit a while ago that the portfolio backtester was made to work with stocks. At the moment, the way it handles futures is unsatisfactory and we keep working on that. We can't get everything done and make it work perfectly right away.

Also, you are saying: why not take care of the backtester first? Other users, in their tern, keep saying: you need to implement this, this and this and as soon as possible. It is clear that it is not possible to satisfy everybody's requests at the same time. And were are fixing hundreds of bugs all the time. It simply takes time to fix everything.

Regards.

Tresor
Posts: 1104
Joined: 29 Mar 2008
Has thanked: 12 times
Been thanked: 53 times

Postby Tresor » 25 Aug 2008


Tresor,

I am not sure how the above problem that fanus helped us identify is related to reversing the positions in backtesting. Anyway, the fact that the logic of reversing the positions is wrong, is a known issue and will be fixed in one of the future versions.

The margin dialogue box has been made to be used in future implementations of MultiCharts.

We did admit a while ago that the portfolio backtester was made to work with stocks. At the moment, the way it handles futures is unsatisfactory and we keep working on that. We can't get everything done and make it work perfectly right away.

Also, you are saying: why not take care of the backtester first? Other users, in their tern, keep saying: you need to implement this, this and this and as soon as possible. It is clear that it is not possible to satisfy everybody's requests at the same time. And were are fixing hundreds of bugs all the time. It simply takes time to fix everything.

Regards.
Hello Marina,

Thank for your reply.

Let me in a few words explain why I bought MultiCharts. I bought MC because:

(i) I could connect RT ascii files to it
(ii) I was amazed by charting capabilities of MC.

The description of MC on your website says '' MultiCharts is a professional technical analysis and automated strategy trading platform for futures, forex, and stock trading, featuring professional charting, advanced analytics, trading strategy optimization, and backtesting.''

When I was buying MC I was a disretionary trader - I didn't care about backtesting and optimization. With time I wanted to dig deeper and check if I could also do some automation to my trading.

At the moment I can confirm that charting is excellent, I can confirm that advanced analytics is okay. The only problem that I have is with strategy optimizing and backtesting.

It is my belief that people buy MC for 3 major reasons (advertised on your website):
1. Professional charting - which is excellent
2. Advanced analytics - which is excellent
3. strategy optimization and backtesting - which needs some fixing

Marina, I do realize that satisfying everybody is practically impossible. I just thought that fixing bugs in Portfolio Backtester prior to a taking care of additional featues would be a natural course for MultiCharts.

Regards

khalaad
Posts: 323
Joined: 07 Jan 2007
Location: Lahore, Pakistan
Has thanked: 64 times
Been thanked: 57 times

Postby khalaad » 25 Aug 2008

Marina,

Having taken a huge loss on my two-year EasyLanguage investment and a small one on the purchase of the ever-more-distant promised MultiCharts Gold, I have no personal interest in how MultiCharts progresses here onwards.

However, for the sake of those still hoping, I must reply. Your answers reinforces my belief TS Support 2008 neither understands nor cares. TS Support 2007 and TS Support 2006 would have both cared and understood.
Other users, in their tern, keep saying: you need to implement this, this and this and as soon as possible.
Do not try to be all things to all people, PLEASE.
It is clear that it is not possible to satisfy everybody's requests at the same time. And were are fixing hundreds of bugs all the time. It simply takes time to fix everything.
TRUE.

Buy you are not promoting MultiCharts on the basis of:
this, this and this
You are promoting MultiCharts for its:

High-defination charting
Dynamic portfolio-level back-testing
True strategy back-testing and optimization
EasyLanguage compatibility
Extensive choice of data feeds
State-of-the-art integrated auto-trading

It makes eminent business sense, and is morally imperative, that in version after version MultiCharts DOES NOT FAIL TO DELIVER ON ANY OF THE ABOVE.
Also, you are saying: why not take care of the backtester first?
Perhaps no one in TS Support notices “back-testing” is the ONLY feature emphasized TWICE?

Or no one REALLY understands and cares?

Khalid

Tresor
Posts: 1104
Joined: 29 Mar 2008
Has thanked: 12 times
Been thanked: 53 times

Postby Tresor » 25 Aug 2008

Marina,

Portfolio Backtester seems to be an important issue to many of us. The issue of fixing it appears on many threads in many posts in different forums. People ask the same questions time and again, they complaint time and again.

I think that if TSS states that it will prioritize its resources to fixing PB, users will contribute with what they think the bugs are / users will elaborate more. This could be a one time effort and the issue might get resolved once and finally to everybody's satisfaction.

Would it be possible and wise to create a sticky thread ''Portfolio Backtester's development - users' suggestions'' where customers could post their ideas on how PB could be changed and could report bugs?

Regards

khalaad
Posts: 323
Joined: 07 Jan 2007
Location: Lahore, Pakistan
Has thanked: 64 times
Been thanked: 57 times

Postby khalaad » 25 Aug 2008

Marina,

You did not address this:
This is why I am harsh: I judge MultiCharts on the basis of claims it makes about its features, and the understanding it offers the users regarding its future development.

In my opinion on both the above counts MultiCharts needs to put its act together.
and this:
They implement features just for the sake of listing it on the website to claim they have a certain feature. If it has bugs, too bad... they are working on the next feature which have "higher priority". In the end you sit with a product with lots of half working features.
and this:
This is the frustration with MC/TS Support. They rather want to implement bells and whistles to make it look pretty instead of focusing on doing the basic things right.
These are matters of greater fundamental importance than the issues you have choosen to write about.

Khalid

khalaad
Posts: 323
Joined: 07 Jan 2007
Location: Lahore, Pakistan
Has thanked: 64 times
Been thanked: 57 times

Postby khalaad » 25 Aug 2008

Tresor,

The claims:

High-defination charting
Dynamic portfolio-level back-testing
True strategy back-testing and optimization
EasyLanguage compatibility
Extensive choice of data feeds
State-of-the-art integrated auto-trading

TS Support make for MultiCharts here

http://www.tssupport.com/multicharts/

must be justified by deeds BEFORE ALL ELSE.

Dynamic portfolio-level back-testing and true strategy back-testing and optimization is poor; state-of-the-art integrated auto-trading is woefully limited.

High-defination charting is excellent; EasyLanguage compatibility is very good; extensive choice of data feeds is quite true.

I think every one buying MultiCharts realises he/she is buying a work in progress.

TS Support must ensure WHATEVER IT IS CLAIMING IS TRUE.

Khalid

fs

Postby fs » 25 Aug 2008

Also, you are saying: why not take care of the backtester first? Other users, in their tern, keep saying: you need to implement this, this and this and as soon as possible. It is clear that it is not possible to satisfy everybody's requests at the same time. And were are fixing hundreds of bugs all the time. It simply takes time to fix everything.
And in my opinion, this is where you go wrong. You implement new features without making sure the ones you have implemented already is working properly. But the worst part is that you appear to be reluctant to fix many of these bugs, because you are working on something new already, or you just give the vague reply "that it will be solved in one of the future releases", which by now has become clear that what you really are saying is "we will probably not fix it, but to stop you from continue to bug us about it, we say we will fix it sometime in the future and hope you forget about it".

In your quest to try and keep everyone happy, it appears that you throw things together without thinking it through properly and you end up with a product which on paper looks like it have a lot of features, but in reality with all the bugs, very few of them are useful. In the end you are not sattisfying anyone since very things are working properly.

If you were still a beta release with a lower price, I think users would have been more accepting of all the issues, but with its current price it is in the same price range as higher end platforms which is far more robust than MC.

khalaad
Posts: 323
Joined: 07 Jan 2007
Location: Lahore, Pakistan
Has thanked: 64 times
Been thanked: 57 times

Postby khalaad » 25 Aug 2008

THE COMMENTS MADE HERE WERE MADE AFTER MARINA POSTED 'OPTIMIZATION AND BACKTESTING BUGS' at 4:57 GMT.

I HAVE VOLUNTARILY REMOVED THEM.

Khalid
Last edited by khalaad on 25 Aug 2008, edited 2 times in total.

User avatar
Marina Pashkova
Posts: 2758
Joined: 27 Jul 2007

Postby Marina Pashkova » 25 Aug 2008

Khalid,

If you want to be heard, you need to stop posting abstract speculations. If you have specific bugs to report or if you have feature suggestions we will be happy to read them. Everything else will be ignored.

khalaad
Posts: 323
Joined: 07 Jan 2007
Location: Lahore, Pakistan
Has thanked: 64 times
Been thanked: 57 times

Postby khalaad » 25 Aug 2008

Marina,

Although I am not quite ready yet to put MultiCharts back on my computer, I consider myself a stakeholder in MultiCharts success.

Now that you have started 'OPTIMIZATION AND BACKTESTING BUGS', I shall write very specific (non-judgemental) comments and suggestions there.

Khalid

Tresor
Posts: 1104
Joined: 29 Mar 2008
Has thanked: 12 times
Been thanked: 53 times

Postby Tresor » 26 Aug 2008

Marina,

Although I am not quite ready yet to put MultiCharts back on my computer, I consider myself a stakeholder in MultiCharts success.

Now that you have started 'OPTIMIZATION AND BACKTESTING BUGS', I shall write very specific (non-judgemental) comments and suggestions there.

Khalid
Khalid,

Thank you very much for commiting to one more effort.

Maina,

Khalid has been helpful to many traders and always wrote MultiCharts up and recommended your product to potential cutsomers: http://kreslik.com/forums/viewtopic.php?t=1224

The above is just one of many examples.

Regards

khalaad
Posts: 323
Joined: 07 Jan 2007
Location: Lahore, Pakistan
Has thanked: 64 times
Been thanked: 57 times

Postby khalaad » 26 Aug 2008

Tresor,

My pleasure. If TS Support will move one step forward, I shall move many.

It is just that the recent months have been so frustatingly disappointing.

Let us hope MultiCharts will facilitate EVERY USER trade better and make higher profits.

Khalid

Tresor
Posts: 1104
Joined: 29 Mar 2008
Has thanked: 12 times
Been thanked: 53 times

Postby Tresor » 26 Aug 2008

Khalid,

Many of us bought MC because we sincerely felt that MC was going to succeed. Everybody wants to be in the winning team :) Recently this feeling faded.

Fs wrote he had sent many e-mails but there was no reaction / denial of obvious issues. He probably was one of many who did so. The frustration grows :(

I hope that one day we will all feel that we are back in the winning team and you will dump SmartQuant / NT / whatever you are using now and will install MC again and continue to support many new traders just as you were you doing till now and -most importatnly - you will continue to use MC to the satisfaction of your equity.

Regards

User avatar
Marina Pashkova
Posts: 2758
Joined: 27 Jul 2007

Postby Marina Pashkova » 26 Aug 2008

Tresor,

I have answered your bug report at http://forum.tssupport.com/viewtopic.php?t=5493

As for fanus, once again, it did take us some time to understand that there really was a problem. Once it happened we aknlowledged the problem and fixed it. What happened in the situation with Fanus was an exception rather than a rule. Also, if fanus still has any issues to report, he is welcome to do so at http://forum.tssupport.com/viewtopic.php?t=5493

Nick
Posts: 496
Joined: 04 Aug 2006
Has thanked: 4 times
Been thanked: 24 times

Postby Nick » 02 Sep 2008

My 2c. (Which I have voiced on numerous occasions over the years)

"Core" functionality needs to be rock solid before moving on to the next "must have" feature. By rock solid I mean fast accurate and reliable. That includes receiving, storing, managing, and displaying data. This gets better over time but I would still not call it "rock solid". Same for the next 'layer' of features. You need to build on rock!

As an aside. I can't imagine people backtesting without proper data management tools (flexible holiday templates, session times, tick filter, roll over, back adjusting etc.)

I guess MC are a small team but that dosen't matter - what is important is you can't leave stuff thats 'kind of working' but buggy or even just 'clunky' while you move on to the next features. Well you can but if it is not fixed in a reasonably timely fashion then you will have dissatisfaction.

For my self I use MC to run a few legacy ELD's that I haven't got round to converting (complicated). I also like EL for quick prototyping simple ideas. I am getting better at doing that in NT but its more complicated. Easy language is the big draw of MC for me. Most ther stuff I'll do elsewhere. I certainly no longer upgrade until things are well settled ( I run 3.0) I just found that the new features did not offset lost trading time. I'd suggest the same to anyone unless they are happy to write off trading time to test.

I also think that modelling stuff on TS is not the best way forwards (beyond offering Easy Language). It is really no longer 'best of breed' in many areas. Thats another issue I guess.

Cheers,
Nick.

Nick
Posts: 496
Joined: 04 Aug 2006
Has thanked: 4 times
Been thanked: 24 times

Postby Nick » 02 Sep 2008

Tresor,

High-defination charting is excellent; EasyLanguage compatibility is very good; extensive choice of data feeds is quite true.

Khalid
Actually there are a few issues with charting still that I have reported in the past. I work around them now. I have discovered a new one recently (aspect ratios of bars getting messed up when you have two widely different time scale data series) but tbh don't have the energy to make the issue 'understood'.

Anyway I guess I was vindicated in my 'MC is not robust enough to autotrade' view though I take no pleasure in this. I sincerely hope that you find a way of achieving what you require.

Cheers.

Tresor
Posts: 1104
Joined: 29 Mar 2008
Has thanked: 12 times
Been thanked: 53 times

Postby Tresor » 02 Sep 2008

"Core" functionality needs to be rock solid before moving on to the next "must have" feature.
Nick, Well said.

User avatar
Marina Pashkova
Posts: 2758
Joined: 27 Jul 2007

Postby Marina Pashkova » 03 Sep 2008

Hi Nick,

Could you please enumerate the features that other programs are better at than MultiCharts? What features are those and in what products?

What do you think is wrong with auto trading? From what I understand, you are still using the 3.0 version. We have released several upgrades, including a release, since then. As a rule, every new version is more reliable than the previous one. A lot of changes have been made in auto trading (as well as other aspects) since then.

What other bugs you think have not been adressed? We have analyzed your requests. For example, we spent quite a lot of time making improvements in the drawings as you suggested.

Regards.

glam_100
Posts: 157
Joined: 14 Jun 2006
Has thanked: 2 times
Been thanked: 9 times

Postby glam_100 » 12 Oct 2008

I think the problem lies in MC's pricing scheme.

Basically MC is a one time investment where a user paid once and then all the bug fix requests is done without any income.

Inevitably this model promotes new features over bug fixes because new features is what draw in new customers. Bug fixes and maintenance becomes a drain on resources that generates no revenue and worse get in the way of the income generating features ("Priorities").

As MC builds up a larger user base this will create a big problem because now you have a huge user base who're not income generating but nevertheless making (rightfully so) a lot of bug fix/maintenance requests.

I don't see how TSSupport can get over with this without charging recuring maintenance fee or monthly fee. Without this the incentive to keep an old customer is simply not there. I'm sure TSSupport would love to keep their old customers and have put in as much effort as possible but the simple fact is that their old customers are not an asset but rather a long term liability. The pricing model is simply broken.

If this go bad enough MC could go down the road becoming the most attractive product for new users (with all the bells and whistles) but get abandoned by all their old customers as it is never really usable for real trading. If this is what they want fine but this is definitely not a long term business model. Revenue generating old customer base is simple a company's biggest asset.

I'm sure many old customers are fair and wouldn't mind a subscription fee for a solid stable product. Many old customers may not agree with this but I think TSSupport should charge $99 fee a month starting from Multichart Gold. Old customers can enjoy a 6 months waived fee and then start paying subscription fee afterwards. This is the only way to create a sustainable product with enough incentives to cater with the ever growing maintenance/bug fix requests.

User avatar
Marina Pashkova
Posts: 2758
Joined: 27 Jul 2007

Postby Marina Pashkova » 16 Oct 2008

Hello glam_100,

I posted a response to your and other users' comments here: http://forum.tssupport.com/viewtopic.php?t=5644

Regards.


Return to “MultiCharts”