Go Back   NinjaTrader Support Forum > NinjaTrader 7 > Platform Technical Support

Platform Technical Support General platform technical support for NinjaTrader 7.

NinjaTrader
Reply
 
Thread Tools Display Modes
Old 11-08-2008, 02:12 AM   #1
ctrlbrk
Senior Member
 
Join Date: Oct 2008
Location: Dallas, TX
Posts: 676
Thanks: 0
Thanked 3 times in 3 posts
Default Benchmark for NT

Guys,

Don't laugh at my terrible C# coding, but I have put together a very simple benchmark for NT.

I wanted to do this with hopes that users of the forum would post their results, along with their system configuration like which processor they have, and then we can compile these results into a table where we can rank NT performance by processor type.

I welcome improvements to the code, I am sure there are numerous ones! To be honest, I am not even sure if the math calculations I am using will realistically represent NT results, so I am hoping to get some official input from the NT team here on this one.

Personally, I am running an Intel E6850 Core 2 Duo overclocked to 3.51 ghz with 8GB of ram, and I use VMware Workstation to 'maximize' my NT backtesting by running more than one instance of NT simultaneously. I am eagerly anticipating two things: a) Intel's Core i7 965 Extreme processor, due out any day, and b) NT 7 which will be multi-threaded.

But, I found myself wondering "how much faster will the Core i7 965 be with Ninja vs my current setup? Or vs a cheaper Core 2 Quad?". This benchmark was written to try and answer that question, by letting the users of this forum participate.

To run the benchmark, just install it and select any symbol in the strategy analyzer, select 1 minute interval, and set the time range to just one market day. The benchmark uses math to decide the speed of the system, it does not depend on how much tick data you feed it.
Attached Files
File Type: zip benchmark.zip (1.7 KB, 203 views)
Last edited by ctrlbrk; 11-08-2008 at 02:31 AM.
ctrlbrk is offline  
Reply With Quote
Old 11-08-2008, 02:16 AM   #2
ctrlbrk
Senior Member
 
Join Date: Oct 2008
Location: Dallas, TX
Posts: 676
Thanks: 0
Thanked 3 times in 3 posts
Default

Please download the attachment, import the script using File -> Utilities -> Import, and then backtest against the strategy "benchmark" using any instrument of your choice, using a 1 minute time frame and one single day for the time period (ie start date today end date today). Be sure to have your Output Window open first!

Results table:

01m 46s: Intel Core i7 920 @ 4.2GHz / 12GB, Seven x64, NT 6.5.1000.10 (sefstrat)
01m 57s: Intel Core i7 920 @ 3.8GHz / 12GB, Seven x64, NT 6.5.1000.9 (ctrlbrk)
02m 04s: Intel Core i7 920 @ 3.56GHz / 6GB, Seven x64, NT 6.5.1000.9 (whitmark)
02m 18s: Intel Core i7 920 @ 3.4GHz / 6GB, Vista, NT 6.5.1000.10 (KrisS)
02m 47s: Intel E6850 C2D @ 3.51GHz / 8GB, Vista x64, NT 6.5.1000.7 (ctrlbrk)
02m 49s: Intel E7300 C2D @ 3.4GHz / 2GB, XP, NT 6.5.1000.9 (dennho)
02m 50s: Intel E8400 C2D @ 4.05GHz / 4GB, XP, 6.5.1000.9 (whittod)
03m 08s: Intel E8400 C2D @ 3.87GHz / 4GB, XP, 6.5.1000.9 (whittod)
03m 14s: Intel Q6600 C2Q @ 3.0GHz / 4GB, Vista x64, NT 6.5.1000.7 (TraderGuy)
03m 19s: Intel Q9450 C2Q @ 3.3GHz / 4GB, Vista, NT 6.5.1000.10 (KrisS)
03m 32s: AMD Athlon 5200+ @ 2.9GHz / 4GB, XP, NT 6.5.1000.10 (eDanny)
03m 55s: Intel T9300 C2D @ 2.5GHz / 2GB, Seven, 6.5.1000.9 (honvl)
03m 58s: Intel E8400 C2D @ 3.0GHz / 4GB, XP, NT 6.5.1000.9 (whittod)
04m 00s: Intel X6800 C2X @ 2.93GHz / 8GB, Vista x64, NT 6.5.1000.9 (jdfagan)
04m 01s: Intel T9600 C2D @ 2.80GHz / 2GB, XP, NT 6.5.1000.8 (heech)
04m 04s: Intel E4600 C2D @ 2.4GHz / 2GB, XP, NT 6.5.1000.9 (whitmark)
04m 04s: AMD Opteron 280 @ 2.4GHz / 8GB, XP x64, NT 6.5.1000.10 (sam028)
04m 07s: AMD Opteron 280 @ 2.4GHz / 1GB, VirtualBox XP, NT 6.5.1000.10 (sam028)
04m 24s: AMD Athlon 3500+ @ 2.2GHz / 3GB, XP, NT 6.5.1000.9 (WhoKnows)
04m 39s: Intel Q6600 C2Q @ 2.4GHz / 8GB, Server 2008, NT 6.5.1000.9 (ctrlbrk)
04m 40s: Intel E6420 C2D @ 2.13GHz / 1GB, RedHat FC9 x86 w/VirtualBox XP, NT 6.5.1000.10 (sam028)
04m 42s: Intel Q6600 C2Q @ 2.4GHz / 8GB, Vista x64, NT 6.5.1000.9 (zoltran)
04m 52s: Intel X3220 Xeon @ 2.4GHz / 2GB, Server 2003, NT 6.5.1000.10 (heech)
04m 55s: AMD Athlon 64 X2 @ 2.10GHz, Vista, NT 6.5.1000.10 (sam028)
05m 00s: Intel Q9300 C2Q @ 2.5GHz / 8GB, Vista x64, 6.5.1000.9 (roonius)
05m 03s: Intel E5345 Xeon @ 2.33GHz / 2GB, Amazon EC2 Server 2003 R2, NT 6.5.1000.10 (heech)
05m 12s: Intel Pentium 4 @ 2.66GHz / 2GB, XP, NT 6.5.1000.9 (whittod)
05m 12s: Intel Pentium 4 @ 3.0GHz / 3GB, XP, NT 6.5.1000.9 (astra)
05m 40s: Intel Pentium 4 @ 2.80GHz / 4GB, XP, NT 6.5.1000.10 (tortexal)
05m 41s: Intel T7300 C2D @ 2.0GHz / 4GB, Vista, NT 6.5.1000.9 (laserdan)
05m 42s: Intel Pentium 4 @ 3.0GHz / 2GB, XP, NT 6.5.1000.10 (rt-trader)
05m 43s: Intel Pentium 4 @ 3.2GHz / 2GB, XP, NT 6.5.1000.9 (whitmark)
05m 51s: Intel T5850 C2D @ 2.17GHz / 3GB, Vista, NT 6.5.1000.10 (darckeen)
07m 27s: Intel Pentium 4 @ 2.41GHz / 1GB, XP, NT 6.5.1000.10 (darckeen)
09m 15s: AMD Athlon XP 2000+ @ 1.67GHz / 512MB, XP SP3, NT 6.5.1000.9 (roonius)
12m 44s: AMD Turion 64 TL-50 @ 1.6GHz / 2GB, Vista x64, NT 6.5.1000.9 (roonius)
Last edited by ctrlbrk; 06-17-2009 at 06:21 AM.
ctrlbrk is offline  
Reply With Quote
Old 03-28-2009, 06:58 AM   #3
whitmark
Vendor
 
whitmark's Avatar
 
Join Date: Nov 2005
Location: Virginia, USA
Posts: 464
Thanks: 6
Thanked 19 times in 13 posts
Default

I applaud your efforts, ctrlbrk, to collect some comparative info on this. To clarify, you only need to use backtest (vs optimize) in the Strategy Analyzer, and since the strategy prints out the elapsed time, make sure you have your Output Window open. Here are some more observations using 6.5.1000.9.

Intel Core 2 Duo, E4600 @ 2.4 GHz, 2GB RAM
Elapsed time: 4 minutes, 4 seconds

Intel Pentium 4, 3.2 GHz, 2GB RAM
Elapsed time: 5 minutes, 43 seconds

Any i7 users care to conduct the benchmark test and report their results?

Regards,

Whitmark
whitmark
NinjaTrader Ecosystem Vendor - Whitmark Development
Last edited by whitmark; 03-28-2009 at 07:00 AM.
whitmark is offline  
Reply With Quote
Old 03-28-2009, 07:35 AM   #4
ctrlbrk
Senior Member
 
Join Date: Oct 2008
Location: Dallas, TX
Posts: 676
Thanks: 0
Thanked 3 times in 3 posts
Default

Thanks whitmark.

I own a i7 920 now. I've updated the second post above to show a results table.

I also added my Q6600 to the list. it's not overclocked, it's a production server (Server 2008). Probably Vista and almost definitely XP with Q6600 would be faster.

Mike
Last edited by ctrlbrk; 03-28-2009 at 08:40 AM.
ctrlbrk is offline  
Reply With Quote
Old 03-28-2009, 08:58 AM   #5
whitmark
Vendor
 
whitmark's Avatar
 
Join Date: Nov 2005
Location: Virginia, USA
Posts: 464
Thanks: 6
Thanked 19 times in 13 posts
Default

Thanks for the update. Btw, the boxes I've tested are running XP. Guess you have a resounding answer to your initial question "a". Now that you have an i7 box, are you finding that the performance differences from running the benchmark to be a good proxy for performance differences one might expect when performing optimization work?

Thanks,

Whitmark
whitmark
NinjaTrader Ecosystem Vendor - Whitmark Development
whitmark is offline  
Reply With Quote
Old 03-28-2009, 09:05 AM   #6
ctrlbrk
Senior Member
 
Join Date: Oct 2008
Location: Dallas, TX
Posts: 676
Thanks: 0
Thanked 3 times in 3 posts
Default

Quote:
Originally Posted by whitmark View Post
Thanks for the update. Btw, the boxes I've tested are running XP. Guess you have a resounding answer to your initial question "a". Now that you have an i7 box, are you finding that the performance differences from running the benchmark to be a good proxy for performance differences one might expect when performing optimization work?

Thanks,

Whitmark
Yes I would say it is pretty dang accurate actually to a "real" optimization job. There is massive headroom here, my C2D @ 3.51GHz usually had both cores around 70% during a backtest/optimization. The i7 920 @ 3.8GHz is barely taxed at around 15% on just one single core.

So, when NT7 comes out, if it truly takes advantage of multi-core, then it should peg all cores @ 100% and if you extrapolate that out it means the optimization could be about 5-6x faster. I guess the big question is, will NT7 do this, or not.

Mike
ctrlbrk is offline  
Reply With Quote
Old 03-29-2009, 09:07 PM   #7
TraderGuy
Certified NinjaScript Consultant
 
Join Date: Apr 2007
Location: Connecticut, USA
Posts: 89
Thanks: 0
Thanked 0 times in 0 posts
Default

Interesting, here is another one.

Intel Q6600, 3.0Ghz (OC), 4 Gig RAM, Vista 64, NT 6.5.1000.7

Elapsed Time
3 Minutes 14 Seconds

CPU was showing 25% - 33% on Task Manager during the test.

Regards,
Guy
TraderGuy is offline  
Reply With Quote
Old 03-30-2009, 04:05 AM   #8
ctrlbrk
Senior Member
 
Join Date: Oct 2008
Location: Dallas, TX
Posts: 676
Thanks: 0
Thanked 3 times in 3 posts
Default

Quote:
Originally Posted by TraderGuy View Post
Interesting, here is another one.

Intel Q6600, 3.0Ghz (OC), 4 Gig RAM, Vista 64, NT 6.5.1000.7

Elapsed Time
3 Minutes 14 Seconds

CPU was showing 25% - 33% on Task Manager during the test.

Regards,
Guy
Thanks for the feedback! Interesting to see two Q6600's side-by-side, and the major difference between stock & overclock. Probably some difference due to Server 2008 vs Vista as well.

I updated the reference post above.

Mike
ctrlbrk is offline  
Reply With Quote
Old 03-30-2009, 11:50 AM   #9
roonius
Certified NinjaScript Consultant
 
Join Date: Oct 2008
Location: Chicago, IL
Posts: 523
Thanks: 0
Thanked 16 times in 11 posts
Default

Intel Core Quad Q9300 (2.5 GHz - not OC); 8GB RAM; Vista Ultimate x64; NT 6.5.1000.9;
5 mins 0 secs

AMD Athlon XP 2000+ 1.67 GHz; 512MB RAM; XP SP3; NT 6.5.1000.9
9 min 15 sec
roonius is offline  
Reply With Quote
Old 03-30-2009, 11:57 AM   #10
ctrlbrk
Senior Member
 
Join Date: Oct 2008
Location: Dallas, TX
Posts: 676
Thanks: 0
Thanked 3 times in 3 posts
Default

Quote:
Originally Posted by roonius View Post
Intel Core Quad Q9300 (2.5 GHz - not OC); 8GB RAM; Vista Ultimate x64; NT 6.5.1000.9;
5 mins 0 secs

AMD Athlon XP 2000+ 1.67 GHz; 512MB RAM; XP SP3; NT 6.5.1000.9
9 min 15 sec
Awesome, thanks for your help roonius.

Mike
ctrlbrk is offline  
Reply With Quote
Old 03-30-2009, 12:23 PM   #11
roonius
Certified NinjaScript Consultant
 
Join Date: Oct 2008
Location: Chicago, IL
Posts: 523
Thanks: 0
Thanked 16 times in 11 posts
Default

On the notebook:

AMD Turion 64 Dual Core TL-50 (1.6 GHz) 2GB RAM; Vista 64; NT6.5.1000.9

12min 44 sec
roonius is offline  
Reply With Quote
Old 03-31-2009, 04:29 AM   #12
ctrlbrk
Senior Member
 
Join Date: Oct 2008
Location: Dallas, TX
Posts: 676
Thanks: 0
Thanked 3 times in 3 posts
Default

Updated, thanks roonius.
ctrlbrk is offline  
Reply With Quote
Old 04-16-2009, 09:38 PM   #13
whittod
Junior Member
 
Join Date: Oct 2008
Posts: 7
Thanks: 0
Thanked 3 times in 3 posts
Default

2mins 50sec
Intel E8400 @ 4.05GHz OC, 1800MHz FSB, 4GB, XP, NT .9

3min 8sec
Intel E8400 @ 3.87GHz OC, 1720MHz FSB, 4GB, XP, NT .9

3min 58sec
Intel E8400 @ 3.0GHz, 1333MHz FSB, 4GB, XP, NT .9

***********************
5min 12sec
Intel P4 @ 2.66Ghz, 533Mhz FSB, 2GB, XP, NT.9
Last edited by whittod; 04-24-2009 at 02:42 PM.
whittod is offline  
Reply With Quote
Old 04-17-2009, 01:20 AM   #14
ctrlbrk
Senior Member
 
Join Date: Oct 2008
Location: Dallas, TX
Posts: 676
Thanks: 0
Thanked 3 times in 3 posts
Default

Updated!

Mike
ctrlbrk is offline  
Reply With Quote
Old 05-01-2009, 01:50 PM   #15
zoltran
Senior Member
 
Join Date: Nov 2005
Location: , Ontario, Canada
Posts: 400
Thanks: 0
Thanked 0 times in 0 posts
Default

4:42 - Q6600 C2Q @ 2.4Ghz ... pretty much the same as yours
Vista64, 8gig ram
zoltran is offline  
Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
building own continuous chart in NT (Data Storage and backfill in NT) Rocket130713 Platform Technical Support 3 08-25-2008 10:36 PM
Give NT Account warning on each NT restart Jim W. Suggestions And Feedback 2 02-22-2008 11:22 PM


All times are GMT -6. The time now is 06:35 AM.

Futures, foreign currency and options trading contains substantial risk and is not for every investor. An investor could potentially lose all or more than the initial investment. Risk capital is money that can be lost without jeopardizing ones financial security or lifestyle. Only risk capital should be used for trading and only those with sufficient risk capital should consider trading. Past performance is not necessarily indicative of future results. View Full Risk Disclosure.

CFTC Rules 4.41 - Hypothetical or Simulated performance results have certain limitations, unlike an actual performance record, simulated results do not represent actual trading. Also, since the trades have not been executed, the results may have under-or-over compensated for the impact, if any, of certain market factors, such as lack of liquidity. Simulated trading programs in general are also subject to the fact that they are designed with the benefit of hindsight. No representation is being made that any account will or is likely to achieve profit or losses similar to those shown.

This website is hosted and operated by NinjaTrader, LLC (“NT”), a software development company which owns and supports all proprietary technology relating to and including the NinjaTrader trading platform. NT is an affiliated company to NinjaTrader Brokerage (“NTB”), which is a NFA registered introducing broker (NFA #0339976) providing brokerage services to traders of futures and foreign exchange products. This website is intended for educational and informational purposes only and should not be viewed as a solicitation or recommendation of any product, service or trading strategy. No offer or solicitation to buy or sell securities, securities derivative or futures products of any kind, or any type of trading or investment advice, recommendation or strategy, is made, given, or in any manner endorsed by any NT affiliate and the information made available on this Web site is not an offer or solicitation of any kind. Specific questions related to a brokerage account should be sent to your broker directly. The content and opinions expressed on this website are those of the authors and do not necessarily reflect the official policy or position of NT or any of its affiliates. 

Vendors along with their websites, products and services, collectively referred to as (“Vendor Content”), are independent persons or companies that are in no manner affiliated with NT or any if its affiliates. NT or any of its affiliates are not responsible for, do not approve, recommend or endorse any Vendor Content referenced on this website and it’s your sole responsibility to evaluate Vendor Content. Please be aware that any performance information provided by a vendor should be considered hypothetical and must contain the disclosures required by NFA Rule 2-29(c). If you are interested in learning more about, or investigating the quality of, any such Vendor Content you must contact the vendor, provider or seller of such Vendor Content. No person employed by, or associated with, NT or any of its affiliates is authorized to provide any information about any such Vendor Content.