Talk:IEEE 1394/Archive 2008

Firewire FUD????

quote: If only raw transfer rate is taken into account, non-optical FireWire is no longer fast enough for the high-end hard drives. Large, sequential reads and writes may have transfer rates which exceed 130 MB/s, which is much more than the maximum transfer rate of FireWire. However, in typical use cases, FireWire still exceeds the sustained and random access data transfer rates of even the fastest high-end hard drives.

I'm going to remove this section. The article at storagereview.com does not mention firewire, and provides only specs and a review of Ultra320 SCSI drives-- "High-end" seems an inappropriate label as well- The article mentions "enterprise-class storage" The latter part seems to be a rebuttle of the first's claim, but again the quoted article is about SCSI drive speeds with no specific mention of Firewire. Both claims seem to be made on theoretical throughput without taking into account drive cache or use in secondary storage/backups as opposed to Operating System media. This section seems to be a about a techie flamewar and not encyclopedic material. PS: I don't consider the relationship between firewire and SCSI technology to be a compelling reason to leave this in. Cuvtixo 21:43, 6 November 2007 (UTC)

The New Seagate 7200.11 Terrabyte Baracuda (i.e. consumer level) drives have a sustained transfer rates (from the platter) of 105MB, too. Tom's Hardware test got 100MB/s http://www.tomshardware.com/2007/11/05/the_terabyte_battle/page5.html That exceeds Firewire 800. That being said, I'm not sure it is relevant to the article. -- KelleyCook 02:15, 7 November 2007 (UTC)

I agree. This is more relevant for an article on external storage solutions.--Anss123 19:47, 7 November 2007 (UTC)

Amusingly, there are vendors who are claiming: "FireWire 800 Data Rates over 115 Megs/second" -http://eshop.macsales.com/shop/hard-drives/RAID/Rack_Mount/FireWire_USB2_RAID_0 which is much faster than the max theoretical rate of FireWire 800! But I would expect such devices to come close to the almost 100 MB/sec theoretical max, since they use versions of RAID (e.g. 0,5) that greater throughput than single spindle drives can. —Preceding unsigned comment added by 76.21.22.240 (talk) 00:22, 16 March 2008 (UTC)

Comparison to USB section

This section gives example speeds for High Speed USB 2.0, but it doesn't provide similar numbers for FW 400 or 800. It makes the assertion "FireWire 800 is substantially faster than Hi-Speed USB" and references a TechTV comparison from 2006, but it makes sense to at least provide similar pre-digested summary numbers as given for USB 2.0. It should preferably provide these same estimates for both FW 400 and 800. 125.197.201.225 (talk) 01:24, 6 April 2008 (UTC)

The article needs a mention of internal FireWire

Apple included internal FireWire ports on some of their G4's[1] and Sonnet Technologies has it on some PCI cards.[2] The article really should have some mention of this, noting that it was intended to encourage the development of internal FireWire devices, but nothing materialized. I'm not certain how to include it in the article. —Preceding unsigned comment added by 4.254.117.60 (talk) 01:17, 22 April 2008 (UTC)

(1.6Gbps) and (3.2Gbps) Firewire

This article should mention the existence of the 1.6Gbps speed and the fact that hardware exists. I don't have time to update it right now. Cmgross (talk) 15:22, 16 April 2008 (UTC)

The article already does mention the 1.6Gbps and 3.2Gbps speeds in the 1394b section. Symwave has just anounced that their new version of their FireWire chip will now enable the already specified the IEEE 1394b S1600 speed. However, they don't yet have a product that you can buy, they are only DEMONSTRATING it. -- KelleyCook (talk) 15:45, 16 April 2008 (UTC)

Wy do we do pretend it doesn't exist. They seem to think they're making parts available to select customers.There is a standard for Firewire 1600 and Fire 3200. There is hardware being announced. Just because the average consumer won't be purchasing it when they run to the store this weekend doesn't mean it is not happening. An online encyclopedia should be about providing current and accurate information. This is especially true in this case since there is a widespread perception that Firewire 1600/3200 are not in development. They're certainly certainly more real than things like UMB. Ultimately, it isn't worth arguing about. I won't update the article if people like it the way it is. In three or four months things will have changed again anyway.Cmgross (talk) 18:58, 16 April 2008

What is difference between what is difference between a proposed technology (Like USB 3), and one that is now in working physical form, but just limited in availability? It will not take long to be in consumer devices. (UTC) --Flightsoffancy (talk) 13:42, 24 April 2008 (UTC)

Wikipedia is to talk about technology, not an advertising service for a chip maker who have pre-announced a product. Individual chipsets aren't mentioned anywhere else on this page, why would this particular be more deserving. The fact that 1.6Gbps products will be coming to market real soon is mentioned in the previous sentence. I'm reverting it again. -- KelleyCook (talk) 18:55, 24 April 2008 (UTC)

Why is this redirected from IEEE 1394

This article should be changed to IEEE 1394; and Firewire should redirect here. This isnt a sales site for Apple.

Wageslave (talk) 20:48, 14 April 2008 (UTC)

I have moved it and all the redirects to Firewire. The article still needs some re-writing to reflect this.--Shniken1 (talk) 01:43, 1 May 2008 (UTC)

Isn't Firewire SCSI?

I thought that Firewire was an evolution in the SCSI protocol, adapting it for serial applications. Even the SCSI article says this. --Dragon695 (talk) 16:29, 9 April 2008 (UTC)

No it is not SCSI. It is a serial interface and I believe it may use some of the SCSI command set (I honestly don't know). However, it is not unique in either regard and that doesn't make it SCSI. Cmgross (talk) 15:13, 1 May 2008 (UTC)

The SCSI article no longer says that Firewire is a form of SCSI. -- Atama^chat 21:24, 16 May 2008 (UTC)

Reversion of two new USB comparisons

I reverted [3] this edit for three reasons:

• It was unsourced WP:OR.
• The first paragraph is not true as their is no clearly defined standard for USB headers either.
• And the second point about cable availability is irrelevant as obviously 1394 beta/bilingual cables are harder to find than their USB counterparts, because the nine-pin interface is much, much less popular than USB 2.x. Monoprice, for example, has them, but mentioning vendors is usually frowned upon. Furthermore USB 3.x will run into the same problem as that group has changed the interface to have an optical connector. -- KelleyCook (talk) 17:06, May 2, 2008 (UTC)

Firewire for DV

The article currently reads:

Almost all modern digital camcorders have included this connection since 1995

But today all (?) consumer camcorders and most professional SD camcorders use USB, if I'm not mistaken. Could someone who knows the history update the article? --Macrakis (talk) 00:05, 23 May 2008 (UTC)

It's not very common for me (as an amateur videographer and editor working for a university A/V dept.) to run into many pro cameras/camcorders using USB 2.0. 1394-400 is the most common connector for mid-grade equipment currently. I imagine pro HD video equipment uses 800/3200 more often than 400. It's a speed issue. You can make video transfers with USB 2.0 (usually found on digital still cams), but every time I can remember doing this, the video is NOT transfered in real time, but instead encoded first into an a/v file and transfered at whatever speed the computer can handle to the HDD for playback. With firewire, not only can you view your capture live, you can edit video right off the camera without having to fill your HDD with uncompressed content. I've never seen USB do this. TechsysPete (talk) 18:32, 22 June 2008 (UTC)

Because this fact is well understood by professionals and amateurs like myself, I will remove the request for a citation from the end of the clause you quoted above. TechsysPete (talk) 18:35, 22 June 2008 (UTC)

Firewire Standards

The 1394 Trade Association has released 1394-2008 which is a single specification for Firewire that replaces all previous versions of the spec. Details at: http://1394ta.org/press/TAPress/2008_0709.html I don't have time to update the article now but it should be tweaked at some point to reflect this, even if it is mostly book-keeping. Cmgross (talk) 15:55, 10 July 2008 (UTC)

Thanks and done. -- KelleyCook (talk) 17:34, 14 July 2008 (UTC)

Target Disk mode

Target Disk Mode anyone? Turns an ordinary ibook into a device (normally the host). How does it pull this off? :) —Preceding unsigned comment added by 166.70.62.200 (talk) 09:04, 28 July 2008 (UTC)

"Target Disk mode" is not part of the 1394 specs but rather something Apple pulled off though the computer's Firmware.--Anss123 (talk) 18:36, 28 July 2008 (UTC)

Cable Length

could someone please add the max cable length? Stephen Charles Thompson (talk) 07:33, 25 October 2008 (UTC)

It was already buried within the text, but I now also put it in the technical specifications where it is more relevant. -- KelleyCook (talk) 17:06, 28 October 2008 (UTC)

old "Standards and versions" list

We should keep them as a historical/time line list when the new 1394-2008 paper that includes all comes out. --Flightsoffancy (talk) 20:02, 29 August 2008 (UTC)

The new standard is published (6 years later!) and I agree. The timeline is still relevant as almost all the consumer implementations still only use 1394a. I have a copy of 1394-2008 and will probably add a blurb about the new stuff, but based on §Q.12 (New features of IEEE Std 1394-2008) it will be a really minor section. -- KelleyCook (talk) 17:10, 28 October 2008 (UTC)

Max Number of Daisy Chained Devices

Everything I can find on IEEE's website (my search results) suggests that the maximum number of devices is 63, rather than 16 as the article suggests. However, I haven't been able to find if this is an addition of 1394a or not and don't want to make an incorrect modification. Can anyone shed light on this? Nate886 (talk) 22:27, 20 November 2008 (UTC)

This has actually been the case since the original release. In fact it is listed in the Scope (1.1c in the 1995 edition and 1.1.1f in the 2008 version). As clause 4.2 (moved to §9.1.2 for 2008) explains, this comes about from figuring the max round trip time for a bus reset/configuration with each repeater adding in the max delay and having a 4.5m cable between each.

Of course, you are correct in that there can be a total of 63 devices on the chain. That is also in the article. If you feel any of this is poorly worded, please edit the article so it is more clear. -- KelleyCook (talk) 01:01, 21 November 2008 (UTC)

iSight on FireWire bus?

I have a Macbook Pro purchased new September 2008. The iSight Camera is not connected to the FireWire bus but to the USB. (According to System Profiler.app) --00:18, 6 December 2008 (UTC) —Preceding unsigned comment added by 99.153.0.248 (talk)

Compaibility of FW400 to FW800?

Technically the protocol is the same and an adapter cable will let you hook the 4 and 6 pin devices with a 9 pin FW controller (computer, etc), but is there any device, audio systems I have read, that do not work with FW800? —Preceding unsigned comment added by Flightsoffancy (talk • contribs) 17:19, 19 December 2008 (UTC)

Kind of a WP:FORUM post, but as long as you have a bilingual FW800 port, then everything should just work. Furthermore, as far as I am aware all the consumer level 1394 Beta ports are bilingual. -- KelleyCook (talk) 19:27, 19 December 2008 (UTC)