nForce3-250Gb: 4-Drive SATA RAID and IDE RAID

Integrated RAID capabilities have become common with most of today's chipsets. Most competing chipsets provide the ability to use SATA RAID 0 (Striping) for Performance or RAID 1 (mirroring) for Data Protection. The nForce3-250 chipset takes these capabilities even further, however, with four drives instead of the two supported by other chipsets. These 4 drives can also be configured as RAID 0+1, which combines both striping and mirroring for speed and data protection.



RAID Level 0+1 Disk Striping and Mirroring


Cross-Controller RAID

The nVidia RAID solution is also unique in supporting both Serial ATA (SATA) and parallel ATA disk devices within a single RAID array. Users don't need to worry about whether hard drives are SATA or IDE - - the nVidia RAID controller treats both IDE and SATA drives the same and allows them to be combined in any way in the RAID controller. A single setup procedure applies to all drives, which makes it possible to use all the storage devices available. Users can also expand the number of drives easily without worrying about whether the drive is IDE or SATA.

On-the-Fly Rebuilds and Spare Disk Allocation

Corporate RAID users are accustomed to RAID arrays with "hot spares" and on-the fly rebuilds, but those RAID 5 features have been expensive additions to a home PC. The nVidia RAID controller adds both these features to a home PC that is running any type of mirroring. If a disk fails, RAID 1 allows continuous operation by taking advantage of the mirrored data copy in the array. The nVidia RAID solution goes a step further and lets a user rebuild a new mirrored copy for the data while the system is up and running, without disrupting user and application access to the data set. Rebuilding on the fly eliminates down time and maximizes protection for critical storage resources.

nVidia RAID 1 also lets users designate spare drives that can be configured as hot standbys, protecting arrays in the event of a disk failure. A shared spare can protect multiple arrays of drives, or a dedicated spare can serve as a hot standby for a particular drive array. The spare disk feature, which offers protection that goes beyond mirroring data, has traditionally been limited to high-end RAID systems. Truthfully, the RAID capabilities that nVidia has built into nForce3-250 will blow away almost any high-end home user. This isn't RAID 5 on a home PC, but the flexibility and data protection capabilities are truly impressive.

nForce3-250Gb: On-Chip Firewall Conclusion
POST A COMMENT

71 Comments

View All Comments

  • arswihart - Monday, April 5, 2004 - link

    In response to #19 sprockkets, sorry this is such a late response, I just checked for responses to my original post. The reason I want Firewire is for Audio Interface purposes, everything from the new Hercules Firewire audio device to Yamaha's MLAN 01X use fireware. Not everything of course, but Firewire is getting very pervasive in pro audio. Reply
  • draven31 - Saturday, March 27, 2004 - link

    note that the S/PDIF spec says that a 'fiber'interface is available... that is a optical S/PDIF. TOSLINK is a type od S/PDIF optical connector. Reply
  • Reflex - Saturday, March 27, 2004 - link

    It really depends on if gaming is your primary use of sound. An Audigy is good for gaming, PEROID. Music affecianado's need not apply. Furthermore, Creative has never really fixed their PCI bus bandwidth issues(possibly will become irrelevant with PCI Express), and can be problematic with other devices due to a crappy ACPI implementation.

    Your diss on the Envy also pretty much ignores its roots in the high end. It is not software audio. It does not do everything that the Audigy does for *gaming* in hardware, but for other functions its all in hardware. It is the ONLY card on the market that not only meets its specs, it exceeds them. The Audigy falls significantly short in several areas(signal to noise, and remember the original Audigy only had 19bit sound despite their 24bit claims, no idea if they fixed that on the Audigy 2 or not).

    For someone serious about sound, an Audigy is not a choice. For a pure gamer, it is an option(although honestly the difference between it and a Envy based solution is negligible). In gaming the Audigy has slightly less CPU utilization and a few more effects, but the sound quality is mediocre at best.

    Personally I do not find that the few effects it adds are worth the downsides of Creative cards. Also, I am more likely to listen to music on my PC than play games, although I do game occasionally. Soooooo....Creative is a poor choice in *my* situation. Your mileage may vary.
    Reply
  • Odeen - Saturday, March 27, 2004 - link

    In the great words of Woody Paige, "How many times do I have to straighten you guys out?"

    Soundstorm:
    Great DSP (which only matters for 3d sound rendering), and has absolutely NO impact on the audio quality, that's the job of the codec chip. Since ALL motherboard manufacturers insist on using the piss-poor Realtek ALC650 chip to do the sound output, the sound quality suffers.

    To see what Soundstorm can REALLY do, check out the Asus A7N266-C, which put 5.1 out on an ACR card that featured a Sigmatel codec, not the ALC650. By moving the analog part of the implementation away from the motherboard, and using quality analog parts, the sound quality (i.e. noise / frequency response / dynamic range)was greatly improved.

    Dolby Digital encoding:
    Don't forget that DD is COMPRESSED. You can't fit six channels of even 16bit/44.1khz audio into a single SPDIF stream. By utilizing DD, you're taking this nice audio generated for you and mp3'ing it on the fly.

    3DSoundSurge.com reviewed the Soundstorm APU and found that the Dolby Digital generated was just six independent streams compressed and "wrapped into" a DD stream. Things like joint stereo weren't utilized at all to share audio information between channels in order to raise the effective bitrate (i.e. if I use 1/2 the bandwidth to describe what's common between two channels, and 1/4 the bandwidth to describe the differences for each channel, then each channel uses an effective 75% bandwidth, instead of just 50%. Ceteris paribus, bitrate = kwalitee. So, DD encoding is a neat idea, but it's a flawed one.
    That said, why not just integrate six or eight digital outputs on a soundcard using VersaJacks? That way, we harness just the 3D audio rendering power of Soundstorm but leave the analog part to external DACs and amplifiers that are chosen by the user.

    It would eliminate the single-cable convinience, but you'd be getting bit-perfect digital output, and it'd be up to the user to pick the DACs and amps he likes. Unfortunately, there don't seem to be any receivers with multichannel digital inputs, but a man can dream of optimal solutions, can't he? :)

    That said, a gamer should still have an Audigy. Since every game out there now uses some form of EAX, you get the best results from using hardware that was designed to support that API, not third-party hardware using someone else's drivers (e.g. Sensaura)

    Speaking of 3D audio rendering, the Via Envy SUCKS. You guys need to realize that Via Envy is just a C-Media 8738 with 7.1 and nice DACs. It's SOFTWARE AUDIO, people, it's AC'97 that sounds a little better than most. It's an eight channel, 24/192- and 24/96-supporting Sound Blaster freakin' Pro! Not that there's anything wrong with that, but, again, all things being equal, playing an EAX-supporting game will have an Audigy2-equipped machine in front, followed by the Soundstorm-equipped machine, followed by a Via Envy-equipped machine.

    Finally, firewire.
    Firewire = good. Chipset-level firewire = gooder. Keep in mind that Firewire has bus-mastering capability, whereas with USB and USB2, the CPU has to handhold every bit going across the bus. Do you really want your shiny new Athlon64 playing crossing guard with USB2 streams, or would you rather have the bits maneuver themselves across independently? Thought so :)

    Chipset-level firewire is good for a simple reason that you only have 133MB/sec maximum theoretical bandwidth. A 400Mb/sec (or 50MB/sec) can eat up to half of your practical PCI bandwidth. Whereas, if it IS integrated, you're only taxing the intra-chipset bandwidth, which is plentiful on A64 boards, and has been plentiful ever since we've gone away from using the PCI bus as the NB/SB interconnect (i.e. the AMD 760 chipset on the AMD side and the Intel BX, which were the last two chipsets to do that).

    WHEW.
    Reply
  • Reflex - Saturday, March 27, 2004 - link

    Whoops, you are correct, I was getting SPDIF mixed up with Toslink cables. My mistake. Heh, I do make those occasionally it seems.

    My point was about the optical Toslink cables, not the digital output itself. However, all that aside, the Soundstorm is still a very low quality integrated sound solution...
    Reply
  • Foxbat121 - Friday, March 26, 2004 - link

    Please check this link for S/PDIF information:
    http://www.mtsu.edu/~dsmitche/rim420/materials/Int...
    Reply
  • Foxbat121 - Friday, March 26, 2004 - link

    #64,

    I don't know what you're talking about. SPDIF is not an optical output. And you don't use optical cable at all. There is also no converter. You ran a coax cable directly from sound card to your receiver's coax input. And it's all digital. There will be no signal loss even if you convert them. However, if you're talking about the different sample rate that causes sound quality issue due to the re-sampling, that is true for most SPDIF ports on board or on sound cards. But that has much to do with the design of the sound card rather than anything else.
    Reply
  • Reflex - Friday, March 26, 2004 - link

    #62: If you are going from an optical output to a coax input, you *are* converting the signal. In a straight optical to optical link, it is being converted first inside the source device and again on the reciever. So yes you are converting the signal. Reply
  • Foxbat121 - Friday, March 26, 2004 - link

    #56,

    While it is true that most people do not base their mobo purchase decision on APU capability, however when it comes to use the PC as HTPC or simply want to play games on your big screen HDTV, the DD real-time encoding plays a big role on chose which mobo to be in your HTPC. Instead of have to connect 3 analog sound wires and pay big $$ to have a receiver to support multi-channel analog input, you can use a SPDIF/Coax digital connection to get all your sound (desktop, game and DVDs) from PC to the HT.
    Reply
  • Foxbat121 - Friday, March 26, 2004 - link

    #58,

    SPDIF is compatible with coax and all you need is a mono mini-jack to RCA adapter so that you can connect it directly to your coax input on the receiver. There is no double conversion needed. I believe that how most people connect their PC to the receiver.
    Reply

Log in

Don't have an account? Sign up now