Jump to content
Sign in to follow this  
Wally

Larrabee

Recommended Posts

Eν αναμονη της Siggraph 2008 μερικα info για την επερχομενη GP-GPU της Intel.

 

CNet

PCPer

TechGage

Hexus

AnandTech

 

Πηζω αυτη τη στιγμη αλλα θα επανελθω με ενα ρεζουμέ...

Share this post


Link to post
Share on other sites

Και μεχρι να ξεβαρεθω...φρεσκα νεα.

Οπως λεγαμε και παλιοτερα,η Intel πιεζει την MS να βαλει το τερας στο επερχομενο κονσολι

Περισοτερα εδω

 

Και οσο για την Siggraph:

Ενδιαφερουσα κουβεντα περι Larrabee + Siggraph

 

PDF paper "Larrabee: A Many Core x86 Architecture for Visual Computing"

 

SIGGRAPH, Larrabee, and The Future of Computing

Share this post


Link to post
Share on other sites

To puzzle αρχιζει να "κολλαει" μαλλον

 

Multicore.info points to an article at EE Times Asia with a partial overview of the approaches to multicore as highlighted by Intel and Microsoft at the recent IDF

 

"Microsoft stressed its vision for placing new layers to its system software stack and point extensions to its .Net environment. Meanwhile, Intel said it plans to have extensions to its x86 instruction set and has shown progress on Ct, extensions to the C++ language with an objective of supporting greater parallelism."

 

Of course, Microsoft is focusing on the plumbing to make parallelism easier to get at and easier to get at advantageously

 

"The underlying software plumbing needs an overhaul before such work can begin. [David Callahan, who leads Microsoft's parallel computing initiative] noted the next systemʼs software will be much more layered into separate elements including new runtime environments that sit in a user space below application libraries and above hypervisors and the core OS kernel."

 

The runtime environments will act as schedulers, working cooperatively with hypervisors that map virtual to physical resources and OSes that manage access to physical hardware. [Ι]“This represents a refactoring of traditional OS services,”[/i] he added.

 

Intelʼs focus is generally at a higher level (applications versus the OS), and at IDF evidently they emphasized their language work

 

"On the language front, Intel talked about Ct, an extension of C++ for multicore processors. The language seeks to automate the job of splitting processing tasks across many cores without the programmer knowing the details of x86 architecture.

 

The language delivers 1.7 to 3.7 times performance speed on code running on four processor systems, according to data shown by Anwar Ghuloum, principal engineer, corporate technology group, Intel. Ct was initially geared toward Intelʼs general purposed Nehalem quad core chips, but is now up and running on its prototype 16-core Larrabee graphics processors."

 

Περισοτερα εδω

Share this post


Link to post
Share on other sites

Θείο,επειδη δεν καταλαβα γρί ο Λαραμπής είναι καλύτερος απο τον Cell ;.......:D

Share this post


Link to post
Share on other sites

Ο Λαραμπης ειναι GPGPU ενω ο Cell CPU ;)

Δηλαδης ο Σελλ(ινος) θελει GPU αφου δεν διαθετει "gpu logic" ενω ο (Χα)Λαραμπης ειναι παληκαρι και τα βγαζει περα και μοναχος του.

Αλλα εισαι τεμπελοπουλος και δεν διαβαζεις τα links που ποσταρω εξου και οι αποριες :P

Share this post


Link to post
Share on other sites

Μαλιστα τα διαβασα τα λινκ και εβγαλα συμπερασματα.

Ο "Larabeeς" βασιζεται

-εχει κλεψει ιδεες-μιμειται

πανω στον "Ceλη".:P

 

Comparison with the Cell Broadband Engine

Larrabee's philosophy of using many small, simple cores has similarities to the ideas behind the Cell processor. However, there are differences in implementation.

  • The Cell processor includes one main processor which controls many smaller processors. In contrast, all of Larrabee's cores are the same, which can be useful for various purposes such as load balancing and task migration.[7]

  • Cell and Larrabee both use a ring bus to communicate between cores.[7]

  • Each of the Cell's SPEs has a local store which is managed explicitly, and they cannot directly access main memory. In Larrabee each core can access all memory through the automatically-managed coherent cache hierarchy.[7].

  • Because of cache coherence, each program running in Larrabee has virtually a large linear memory just as in traditional general-purpose CPU. In contrast, an application for CELL should be programmed taking into consideration limited memory footprint (256KB) of the local store associated with each SPE, making it more difficult to develop programs.

  • Cell uses DMA for flexible data transfer to/from on-chip local memories; whereas Larrabee uses special instructions for cache manipulation (notably cache eviction hints and pre-fetch instructions), which have a notable advantage in that we can maintain general [memory hierarchy|cache coherence] while increasing performance for e.g. rendering pipelines and related stream-like computation.

 

Share this post


Link to post
Share on other sites

Τι διαφορες βρε Aardy?Του ριχνει στη κεφαλα απο οτι φαινονται και στα bullets που ποσταρες :)

 

Kαι αν λαβουμε και υποψιν το γεγονος οτι το scalability του ειναι αναιμακτο (σε αντιθεση με το Cell οπου το κοστος και η πολυπλοκοτητα ανεβαινουν εκθετικα οσο προστιθενται core-ς...) και η προσθεση cores εχει, λενε ,linear αποτελεσματα (οχι οπως τα crossfire και τα λοιπα κολπα) τοτε..."οποιος το λαβει,τον κυριον ειδε" :D:D

 

Ενα απο τα πολλα που δεν μπορω να καταλαβω με το φτωχο μου μυαλο ειναι πως η Intel θα πολεμησει την καταναλωση (αν μαλιστα παει στο φημολογουμενο 32cores-2Ghz setup) ωστε να της επιτρεπει να το βαλει σε κονσολα...

Share this post


Link to post
Share on other sites

To Enemy Territory: Quake Wars re-done ως raytraced απο την Intel.

Βλεπω τo Larrabee να πλησιαζει (ευχαριστα) επικινδυνα! Αντε θελουμε καινουργιο ΧΒΟΧ τελη 2009 ρεεεεεε!

 

"Intel released the article 'Quake Wars Gets Ray Traced' (PDF) which details the development efforts of the research team that applied a real-time ray tracer to Enemy Territory: Quake Wars . It describes the benefits and challenges of transparency textures with this rendering technology. Further insight is given into what special effects are most costly. Examples of glass and a 3D water implementation are shown. The outlook hints into the area of freely programmable many-core processors, like Intel's upcoming Larrabee, that might be able to handle such a workload."

Share this post


Link to post
Share on other sites

Βρε τεμπελχανοπουλε δεν διαβασες το PDF?

 

picture.php?albumid=47&pictureid=5413

Share this post


Link to post
Share on other sites

Δημιουργήστε ένα λογαριασμό ή συνδεθείτε για να σχολιάσετε

You need to be a member in order to leave a comment

Δημιουργία λογαριασμού

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

  • Recently Browsing   0 members

    No registered users viewing this page.

×