Tianli
Posts:1
 |
| 05/23/2008 8:15 PM |
 Quote
 Reply
 Alert
|
I have some questions on the specs ver 1.1 (NYSE OpenBook Ultra Customer Interface Specification - ver1.1.pdf).
1. The specs mentions "NumPricePoints" field several times (e.g., page 28), but I can't find this field in any of the messages.
2. It seems all messages are fixed length, for example, Full Update Message - 60 bytes; Delta Update Message - 62 bytes. But in the format table the MsgSize is a function of "NumPricePoints" field (back to Question 1...). Are they fixed or variable length? If variable, what fields are repeated?
3. In the PDP header, the "NumBodyEntries" field is,
"The number of times the message body repeats in the message. For example, if the body consists of a field (named Volume) and the “NumBodyEntries” field is 2, the number of bytes in the message body will be 8."
Please what is the "message body" here? What fields exactly are repeated and where?
4. The "LinkFlag" field in the PDP header, is it the only field used to identify messages that span multiple packets? That is, something like PSP - packet sequence number in the Openbook Real-time feed.
5. The SymbolSeqNum field "increases serially but not monotonically". Does this mean there can be gaps within the symbol's message stream? What's the use of it if the symbol seq number is not continuous? Obviously it can't be used to detect gaps like SSN in Openbook Real-time if it's not continuous.
Thanks.
|
|
|
|
|
Ariel
Posts:170
|
| 05/30/2008 3:44 PM |
Quote
Reply
Alert
|
a new spec has been uploaded (ver 1.2). It should clarify what the NumPricePoints means as the price point fields can repeat in a message. The rest of the fields in the message are fixed.
The linkflag is set to one in the case of a retransmission request, when the a full book update or messages requested can span multiple packets before the requestis fulfilled.
The SymbolSeq Num is increased serially and monotonically the spec will be updated |
|
|
|
|
developer (guest)
|
| 06/04/2008 12:23 PM |
Quote
Reply
Alert
|
| Assuming one was only interested in the inside price level, what advantages/differences are offered by nyse openbook ultra over the nyse bbo feed? |
|
|
|
|
developer (guest)
|
| 06/04/2008 12:25 PM |
Quote
Reply
Alert
|
| Since the feed ultimately represents limit orders, why not make it the same as nyse's arcabook? It would save us programmers quite a bit of development time =)... |
|
|
|
|
Ariel
Posts:170
|
| 06/04/2008 4:34 PM |
Quote
Reply
Alert
|
| The OpenBook Ultra format will be ultimately the common format across all our markets. It started as the ArcaBook format but was changed after customer feedback. Ultra doesnt contain only Limit orders, it contains displayed crowd and specialist interest as well. |
|
|
|
|
developer (guest)
|
| 06/05/2008 10:26 AM |
Quote
Reply
Alert
|
| Thanks Ariel, though I must admit I'm at a loss as to why crowd and specialist interest couldn't be represented as virtual limit orders under the same protocol. Any color regarding the NyseBBO feed? Will the inside price levels match between the 2 feeds and have similar speed of dissemination at this time? |
|
|
|
|
Turambar (guest)
|
| 06/05/2008 6:34 PM |
Quote
Reply
Alert
|
Hello:
When "Slow Quote" status is set, the NYSE stops sending quotes to the SIAC CQS.
I want to know whether or not the NYSEOpenBook releases quotes even in the event of a "Slow Quote" status.
Regards:
Turambar. |
|
|
|
|
Ariel
Posts:170
|
| 06/06/2008 9:15 AM |
Quote
Reply
Alert
|
Hi Turambar,
Just to correct your statement about "slow quotes". When a slow quote status is set, we send A, B, H quote conditions to CQS to indicate that the NYSE Market has switched off electronic Trading for a max 5 seconds to allow the specialist to inject more liquidity into the market. The same occurs when an LRP is hit, we send E, F or U quote condition to CQS. Either way the quote is always sent to CQS. OpenBook will update regardless of the quote or sale conditions. It is even driven, so every Order, Cancel, Execution will result in an OpenBook message. |
|
|
|
|
Ariel
Posts:170
|
| 06/06/2008 9:18 AM |
Quote
Reply
Alert
|
| To answer the prior question on the Best Quotes feed: "Any color regarding the NyseBBO feed? Will the inside price levels match between the 2 feeds and have similar speed of dissemination at this time?" The inside (or Top of Book) in OpenBook Ultra will match the Best Quote sent out in the August. Until then Best Quotes contains slightly additional interest that is not on OpenBook Ultra yet. Speedwise, the top of book on Ultra will always be faster because its a direct route out. |
|
|
|
|
Turambar (guest)
|
|
guest (guest)
|
| 06/08/2008 8:43 AM |
Quote
Reply
Alert
|
| Is there any plan to include the market orders in the feed? |
|
|
|
|
Lydia
Posts:68
|
| 06/09/2008 9:19 AM |
Quote
Reply
Alert
|
| Thank you for your inquiry. We are working on your question and will respond shortly. |
|
|
|
|
Kevin
Posts:21
|
| 06/09/2008 12:31 PM |
Quote
Reply
Alert
|
There are no plans to show market orders in OpenBook, it consists of limit orders.
Kevin |
|
|
|
|
AL (guest)
|
| 06/13/2008 3:35 PM |
Quote
Reply
Alert
|
Missing delta update messages…
While verifying my NYSE OB Ultra implementation I have noticed many gaps in the delta-update messages, for example – looking at the AIG delta updates I identify several gaps every few minutes (about 2% of the updates are missing).
I’m not missing any message, and using wire-shark I have verified that while I’m not missing any message the updates are really missing on the wire (record file is available).
I have two questions:
1. To NYSE team – who can I call to verify feed integrity?
2. To the users community – since this is a new product - is there anybody who has successfully integrated to the feed and has no missing updates?
Thanks.
|
|
|
|
|
Tianli
Posts:1
|
| 06/13/2008 4:40 PM |
Quote
Reply
Alert
|
| We've noticed the same thing, there can be gaps in Symbol Seq Num even when the Line Seq Num is continuous, i.e., no packet's missing. According to the spec this should never happen. |
|
|
|
|
Lydia
Posts:68
|
| 06/16/2008 10:12 AM |
Quote
Reply
Alert
|
AL and Tianli,
Thanks for your inquiries. I would like to let you that the issues you identified are being corrected in a software update going online on June 30. The spec is correct and the symbol sequence numbers do increase sequentially.
Regards,
NYX Data Team |
|
|
|
|
guest (guest)
|
| 07/08/2008 8:42 AM |
Quote
Reply
Alert
|
| I just wanted to confirm that the gaps in delta messages and NULL reason codes are not yet live. I am still seeing both these issues in the live feed. |
|
|
|
|
Jay (guest)
|
| 07/11/2008 4:42 PM |
Quote
Reply
Alert
|
What exactly is sent in response to a Book Refresh Refresh? Is it just a Full Update message (228)?
Also, according to the section 4.2 of the OpenBook Ultra 1.4 spec, "Any Update Message can span multiple messages and/or multicast packets." Is this related to the "LinkFlag" in the header? The documentation on that fields says "This field applies only to Refresh Retransmissions...For all other messages...this field is always set to 0." How are these two points reconciled? If the LinkFlag is not used, then what does it mean for a message to span multiple messages or packets? Does it simply mean that it may take more than one Update Message to provide all of the price points for a given symbol?
Thanks |
|
|
|
|
Matthew (guest)
|
| 07/17/2008 9:29 PM |
Quote
Reply
Alert
|
[quoteΦ. The "LinkFlag" field in the PDP header, is it the only field used to identify messages that span multiple packets? That is, something like PSP - packet sequence number in the Openbook Real-time feed.[/quote]
If the linkflag only identifies packets for refresh retransmission how do you identify how many packets represent a single full update message for a single symbol in the Openbook Ultra datafeed that spans multiple packets? |
|
|
|
|
ztang (guest)
|
| 07/18/2008 9:18 AM |
Quote
Reply
Alert
|
| Is it possible to get quote condition from this feed? Thanks. |
|
|
|
|