G.Skill Falcon SSDs

[taltamir]

Right now the competition is selling drives for about the same amount, or even lower than OCZ. I don't claim to know anything about margins, but over a month worth of consulting is worth of millions of dollars in the business field, this is not even including the cost of the original costs of collaboration with Indilinix.

And yet most of the consulting was by anandtech, for free.

All of this hinges of course, on how long OCZ plans to sell this product in order to help recoup all these costs. Like I said, if Indilinix's controller gets cheaper to make eventually, OCZ will take a hit because it will be paying more than for the controller than its competitors (since you are typically locked to a contract price).

And if their contract department has half a brain, their contract stipulates for such cases.

If all OCZ wanted was for "market penetration" of their brand name, then its a pretty ballsy move and may or may not end up coming to bite them in the ass due the current size of the SSD market--although its possible they are hoping for spillover effects to their other product lines. Otherwise, my guess is, they hope they start selling a HELL of a lot of drives before they end up like Microsoft or Sony, paying an exorbitant amount of money for G70/R500 technology 3 years later to help nvidia/ati pay for R&D costs for GT300/RVV870.

Wanted/planned vs result are different things. They could have a positive result like you said that they didn't plan for or expect.

 

[alcoholbob]

Originally posted by: taltamir
And yet most of the consulting was by anandtech, for free.

I have not heard Anandtech taking any credit for developing 1199, 1275, and 1370 firmwares. Proof?

And if their contract department has half a brain, their contract stipulates for such cases.

So it takes half a brain to outmatch Microsoft or Sony's legal department on one front, but apparently it would take too much from this same half-brain to achieve exclusivity. Gotcha.

 

[Idontcare]

Originally posted by: Astrallite

Originally posted by: taltamir
And yet most of the consulting was by anandtech, for free.

I have not heard Anandtech taking any credit for developing 1199, 1275, and 1370 firmwares. Proof?

FWIW

Disappointed, I went back to OCZ

OCZ was worried. The last time I reviewed one of their SSDs I was truthful about it, and it hurt their sales considerably. Customers were returning drives, and to OCZ?s credit, they stepped up and even accepted some returns themselves - something that most manufacturers wouldn?t have done. Regardless what they had told me, there was some admission of fault there. Those JMicron drives were nothing short of crap.

As soon as OCZ started getting word that I wasn?t pleased with Vertex, they went into a state of panic. These drives all do very well in synthetic HDD tests like HDTach and ATTO, that?s generally all they?re reviewed in, so that?s all they?re tested in. But now OCZ was hearing that the Vertex wasn?t passing some of my tests and they had no idea what it was failing or why.

I tend to do a good job of keeping what tests I run secret until the review is published, so there isn?t any unfair optimization. I purposefully introduce new tests to our performance suites to help keep manufacturers honest and optimizing for real world usage scenarios rather than specific benchmarks. OCZ had no idea what I was running, but they knew that the Vertex wasn?t doing well.

Summit on the other hand was performing just fine, but that?s an expensive drive. Vertex was supposed to be good, it should?ve been good, there?s no reason for it to be performing this poorly. I ran the infamous iometer test to see what was going on:

Iometer 4KB Random Writes, IOqueue=1, 8GB sector space IOs per second MB/s Average Latency Maximum Latency
Original Pre-release OCZ Vertex 20.7 0.08 MB/s 48.2 ms 484.5 ms

How on earth is this acceptable at all? Average latency of 48.2ms and a maximum latency as bad as the Apex and G.Skill Titan drives? I?ve heard some SSD vendors dismiss the iometer results but let me caution you against that. What these numbers are telling us is that on average, when your OS goes to write a 4KB file somewhere on your drive, it?ll take nearly 50ms. That?s 4.5x longer than a 5400 RPM 2.5? notebook drive and that?s the average case. What part of that sounds acceptable? Anyone who tells you otherwise is delusional.

I thought for sure that the drive was broken and that we?d made no progress since last fall. But the drive hadn?t launched yet, while there were glowing reviews of it, no one had wasted any money. I wrote an email to Ryan Petersen, OCZ?s CEO. I described my findings and told him that while the Vertex?s performance was better than any of the JMicron solutions, it was unacceptable for anything other than perhaps extremely light, single-tasking usage.

I told him it sucked. He said that wasn?t fair. We argued over email but he came back and asked me what I needed to see to make the drive better.

I told him I?d need an average response time in the sub-1ms range and a max latency no worse than Intel?s 94ms. I didn?t think it would be possible. I was prepared for OCZ to hate me once more. He told me to give him a couple of days.

http://www.anandtech.com/stora...owdoc.aspx?i=3531&p=20

 

[faxon]

Originally posted by: Astrallite

Originally posted by: taltamir
And yet most of the consulting was by anandtech, for free.

I have not heard Anandtech taking any credit for developing 1199, 1275, and 1370 firmwares. Proof?

someone needs to read the articles Anand himself posts more often. He was the one who was responsible for telling OCZ the first firmware version on the vertex sucked (and it did), and getting it fixed in a matter of days so as to get a retail drive to him in time for the review

 

[alcoholbob]

Originally posted by: faxon

Originally posted by: Astrallite

Originally posted by: taltamir
And yet most of the consulting was by anandtech, for free.

I have not heard Anandtech taking any credit for developing 1199, 1275, and 1370 firmwares. Proof?

someone needs to read the articles Anand himself posts more often. He was the one who was responsible for telling OCZ the first firmware version on the vertex sucked (and it did), and getting it fixed in a matter of days so as to get a retail drive to him in time for the review

Okay, then "guy", for someone (you, apparently) whose read the articles, wheres the proof that Anand had anything to do with 1199, 1275, and 1370? It seems like my reading comprehension is lacking. Please, don't be shy. I REALLY hope you have something to say since I trust you aren't the shameless sort to just type-and-hide.

Originally posted by: Idontcare

FWIW

Well some people disagree about Anand's actual "participation" in this matter regarding the pre-release firmware to 0122 (although he liberally takes credit here).

Since Anand's review 3 other firmwares were released.

1) 1199 - Attempt at faster sequential numbers than 0122. However, 1199 lacked stability and often crashed/lockedup during sequential writes.

2) 1275 - Addressable drive space was reduced by 2-3% to act as a buffer.

3) 1370 - same as 1275, except its hard coded to work only with the OCZ Vertex. Beta TRIMWIPER application will only run if this firmware is present. Read/Write performance is more consistent across multiple stripe sizes.

Now I haven't seen Anand take any credit for these firmwares. OCZ has been very incensed that the firmwares they've developed are hot-flashable to OEM Indilinix products which is why 1370 is only flasable if it has a OCZ Serial number (and by that extension, Trim wiper)

Now, since our intrepid fellows Taltamir and faxon believe Anand should be paid for helping OCZ develop 1199, 1275, and 1370, my answer is this: If Anand really had his knees this deep in OCZ affairs to be their software engineer for 3 straight months and is doing this pro bono, then he's probably tea-bagging the CEO's daughter (no pun intended).

 

[taltamir]

i wouldn't say anand takes CREDIT for those... anand talked to OCZ and gave consultation on what the drive SHOULD do, which resulted in OCZ going to the chip maker, whose engineers adjusted the firmware.

But we were discussing consultation, and telling someone "this sucks, for it to be a quality product it needs this and that" is the very definition of consultation services.

Also, I don't recall mentioning a specific firmware version, definitely not ones created after anand's review.

 

[Idontcare]

Originally posted by: Astrallite

Originally posted by: Idontcare

FWIW

Well some people disagree about Anand's actual "participation" in this matter regarding the pre-release firmware to 0122 (although he liberally takes credit here).

Since Anand's review 3 other firmwares were released.

Yeah that's why I lead into it with the "for what's its worth" caveat.

Again, FWIW, in the IP world the successive works involving the three follow-on firmwares could be considered as "derivative works" whose origins are fundamentally dependent on the "work" Anand was involved with.

There is a difference between being a consultant versus being an engineer. Consultants provide feedback and guidance/direction to the engineer (or more typically the engineer's boss/project manager) without which the engineer would not have necessarily stumbled upon the solution that better maximizes the market opportunity. I've been both, engineer and consultant, in my lifetime. Consultant was the better choice, $1k/day is way better than $0.4k/day, and they both provide value-add in their own ways to the formation of an overall more sellable product.

To say Anand's feedback (a form of pro-bono work, let's not begrudge him all the credit here) was not pivotal in the evolution of the Vertex is, to put it gently, simply ignorant (I hate using that word, it just comes across so harsh and derogatory which is not what I mean to imply here, but for lack of a better word I am compelled to use it here) of the development and productization process that goes into just about any commercial hardware involving engineering, consulting and marketing.

At the same time to attempt to make the argument to the other extreme and posit that without Anand's feedback that all of the Vertex offerings (and subsequent firmware advancements) would have never existed and thus he provided millions in free consulting work is a tad overly generous and equally ignorant of how productization works in general.

It's natural for all parties to overstate the criticality of their impact to the overall product, it is very much a self-promote or fade-away world out there when it comes to individual career development...but its pretty much multiple-industry wide standards for consultants to provide informations that lead to derivative works in which no one (save a few egotistical engineers) would begrudge the consultant their kudos for bringing some degree of value to the overall product. (just as most consultants who aren't inherent asshats would equally not begrudge the engineers their kudos for their portion of the value added to the product)

The NIH syndrome in general selectively targets engineers for some reason, I was infected for while too. Time, age, and maturity cured it eventually, thankfully.

 

[dighn]

First review that I've found: http://www.tweaktown.com/revie..._state_disk/index.html

Looks good so far. The write latency looks bad in that chart, but is probably on par with the vertex judging from how much faster in that area the x25-m is compared to the vertex from the anandtech review.

 

[VaultDweller]

Originally posted by: Astrallite
wheres the proof that Anand had anything to do with 1199, 1275, and 1370?

Where's the proof that OCZ had anything to do with it? Indilinx engineers the firmware, not OCZ. You're assuming that OCZ is providing extensive and costly consulting services to Indilinx for free - what are you basing this assumption on?

 

[Idontcare]

Originally posted by: dighn
First review that I've found: http://www.tweaktown.com/revie..._state_disk/index.html

Looks good so far. The write latency looks bad in that chart, but is probably on par with the vertex judging from how much faster in that area the x25-m is compared to the vertex from the anandtech review.

Where's the small file random write benches? Did I miss them?

 

[taltamir]

Originally posted by: Idontcare

Originally posted by: Astrallite

Originally posted by: Idontcare

FWIW

Well some people disagree about Anand's actual "participation" in this matter regarding the pre-release firmware to 0122 (although he liberally takes credit here).

Since Anand's review 3 other firmwares were released.

Yeah that's why I lead into it with the "for what's its worth" caveat.

Again, FWIW, in the IP world the successive works involving the three follow-on firmwares could be considered as "derivative works" whose origins are fundamentally dependent on the "work" Anand was involved with.

There is a difference between being a consultant versus being an engineer. Consultants provide feedback and guidance/direction to the engineer (or more typically the engineer's boss/project manager) without which the engineer would not have necessarily stumbled upon the solution that better maximizes the market opportunity. I've been both, engineer and consultant, in my lifetime. Consultant was the better choice, $1k/day is way better than $0.4k/day, and they both provide value-add in their own ways to the formation of an overall more sellable product.

To say Anand's feedback (a form of pro-bono work, let's not begrudge him all the credit here) was not pivotal in the evolution of the Vertex is, to put it gently, simply ignorant (I hate using that word, it just comes across so harsh and derogatory which is not what I mean to imply here, but for lack of a better word I am compelled to use it here) of the development and productization process that goes into just about any commercial hardware involving engineering, consulting and marketing.

At the same time to attempt to make the argument to the other extreme and posit that without Anand's feedback that all of the Vertex offerings (and subsequent firmware advancements) would have never existed and thus he provided millions in free consulting work is a tad overly generous and equally ignorant of how productization works in general.

It's natural for all parties to overstate the criticality of their impact to the overall product, it is very much a self-promote or fade-away world out there when it comes to individual career development...but its pretty much multiple-industry wide standards for consultants to provide informations that lead to derivative works in which no one (save a few egotistical engineers) would begrudge the consultant their kudos for bringing some degree of value to the overall product. (just as most consultants who aren't inherent asshats would equally not begrudge the engineers their kudos for their portion of the value added to the product)

The NIH syndrome in general selectively targets engineers for some reason, I was infected for while too. Time, age, and maturity cured it eventually, thankfully.

eloquently put

 

[dighn]

Originally posted by: Idontcare

Originally posted by: dighn
First review that I've found: http://www.tweaktown.com/revie..._state_disk/index.html

Looks good so far. The write latency looks bad in that chart, but is probably on par with the vertex judging from how much faster in that area the x25-m is compared to the vertex from the anandtech review.

Where's the small file random write benches? Did I miss them?

Page 7. Looks like 64 KB random access test (from the screenshot).

I was kinda looking forward to this, but just saw a good deal (for Canada anyway) on the X25-M 80 GB today and jumped on that instead.

 

[Idontcare]

Originally posted by: dighn

Originally posted by: Idontcare

Originally posted by: dighn
First review that I've found: http://www.tweaktown.com/revie..._state_disk/index.html

Looks good so far. The write latency looks bad in that chart, but is probably on par with the vertex judging from how much faster in that area the x25-m is compared to the vertex from the anandtech review.

Where's the small file random write benches? Did I miss them?

Page 7. Looks like 64 KB random access test (from the screenshot).

I was kinda looking forward to this, but just saw a good deal (for Canada anyway) on the X25-M 80 GB today and jumped on that instead.

Ah, thanks! I was thinking more of the traditional 4KB "small" file tests. 64KB never really highlighted the problems with Jmicron controllers, and I doubt it will yield insight into any concerns we might have for the Falcon as well.

Having said that, and garnering myself a well-earned Captain Obvious award, it begs the question why are reviewers intentionally avoiding doing the small file (4KB) random write tests when it is obvious to everyone that those are the benches that separate the men from the boys in the world of SSD's.

When reviewers appear to gloss over the details like this and head straight for the "zomg the 1MB sequential reads are teh bombz" conclusions I come away feeling like the reviewer is just an extension of the marketing dept of the manufacturer or they really are ignorant and have no business doing SSD reviews in the first place. Running a 4KB random write bench couldn't be any easier, crystaldiskmark ftw, so what plausible excuse does any reviewer have for not including benches in their review?

 

[alcoholbob]

Originally posted by: VaultDweller

Originally posted by: Astrallite
wheres the proof that Anand had anything to do with 1199, 1275, and 1370?

Where's the proof that OCZ had anything to do with it? Indilinx engineers the firmware, not OCZ. You're assuming that OCZ is providing extensive and costly consulting services to Indilinx for free - what are you basing this assumption on?

OCZ is testing the product in the wild by having the first product with the controller. Vertex customers made hundreds of complaints in the first few weeks of release, and several of the OCZ reps were able to reproduce these errors themselves after being notified. They stated they were forwarding these results to Indilinix to have them review and find solutions. Do I have any *proof* Indilinix did not find these problems out in their labs without input from OCZ or their customers? Can I prove that OCZ wasn't lying and their company reps weren't just blowing smoke and sitting on their asses and never even told Indilinix there was a problem? No. Everything I say is based on inference. How do you want me to prove it?

I *believe* PC Perspective were the ones who found out about the Intel 8610 firmware woes because their article infers it, and many online publications give them credit for it. If you challenged me to prove it, can I? Can I prove Intel did not find out this problem by themselves in their own labs prior to outside input, that PC Perspective took credit and lied about Intel's response, and that Intel wasn't already in the process of rolling out their new firmware? I can't prove that.

To me, these things seem "obvious"--if you went to college, this is called "folk psychology science." If a kid steals food, and happens to be starving to death, can I PROVE that he stole because he was starving to death? No, I can't. But most things in life fall under folk psychology and can't be bottled up in physics lab.

Okay "dude", you want to add your input? Then prove Anand had anything to do with these firmwares. Like I said, I have reading comprehension issues, and going through his articles I can't find a single reference to his involvement in 1199, 1275, and 1370. But I COULD BE WRONG. If you demand proof then back up your statement. I challenge you to respond to this. If you don't, then you are a straight up coward.

 

[n7]

This is getting off topic.
Who deserves credit for the Vertex's firmware can be discussed in a new thread, or PMs.
Thanx.

n7
Memory/Storage Mod

 

[Idontcare]

Guys I've got no issue with tenants of the conversation going on in this thread, but the tone is kinda concerning me.

Can we just kinda cool it down a little, focus more on the facts as we know them, seek some common ground on the parts of the backstory that we will never know as fact but just speculation, and ask ourselves why it even matters to us or the rest of the world?

I see every post in this thread as having some elements of truth and plausibility to them, no one is saying anything that seems overly implausible or illogical or refutable. Not one of us here actually knows any of the players personally to have confidence that even as second-hand information we know what the story is between Anand and OCZ.

So let's speculate, and debate, and openly air our opinions, but lets due it without building ourselves up to a point of brinkmanship is all I am asking/pleading, as a fellow poster who enjoys reading everyone's contributions to this thread and forum.

edit: mod beat me to it while I was writing my post

 

[sxr7171]

Originally posted by: Idontcare

Originally posted by: skriefal

Originally posted by: Astrallite
If for all they did, they only got the Indilinix controller for cheaper, that's a huge loss for them. Eventually Indilinix's manufacturing costs will go down, and OCZ will keep paying the same price for the controllers while new partners will be paying for less.

Loss? Not really. Again, they get to influence the design of the controller such that it works how they and their customers want, rather than accepting a controller that was implemented to meet the demands of some other OEM. There's value in that. (Real value or perceived value? That's another discussion.) And they were first to market with the Indilinx controller -- there's definitely value there.

None of us can say whether the price per controller is or will be an issue; their contract may already allow for adjustments based upon the cost paid by other vendors.

Intel seems to do just fine keeping their controller all to themselves...as do all the other harddrive manufacturers.

I'm just surprised with OCZ's business model in this area, trying to eek out margins on a product that essentially cannot be differentiated from the other rebranded generics on the market is a brutal way to try and turn a profit.

Imagine if your business model was to buy Intel CPU's, label them as your own, and resell them to compete with Intel's otherwise identical CPU's. You could make some money doing it, buy in bulk getting a volume-discount then resell at MSRP, plus you'd probably carry the warranty yourself so you'd buy OEM from Intel instead of retail. But dam your margins are going to be slim-to-none.

Firstly Intel SSDs are resold by Kingston.

http://www.fadfusion.com/selec...tem_number=10068600106

Second what is OCZ after all? They are a marketing and support organization. Not a supplier or designer. Starting from scratch what kind of controller could they design?

Nothing.

It's very clear that OCZ does not design anything but ask for a few tweaks on an existing design. That and they nowhere near enough to afford to own the Indilinx design. Even if they did own it, they themselves would be licensing it out to make it worth its cost. This isn't Intel that can hold onto its IP and still sell enough to make it worth their while.

 

[AbRASiON]

Is the G.Skill Falcon going to be supported by G.Skill aswell as OCZ seem to be promoting the Vertex?
I notice the G.Skill is fairly cheap for 128gb over at newegg (about 40$ less than the OCZ)
Will OCZ firmware work on it, if not does G.Skill address things as frequently as OCZ?

Refuse to take a risk at 450$ AUD

 

[alcoholbob]

Chances are the firmware is functionally identical but things don't always work out with cross-brand flashing, and if problems exist the last thing you want to do is send it in for RMA and your gskill drive comes up as an OCZ Vertex.

 

[Idontcare]

Originally posted by: Astrallite
Chances are the firmware is functionally identical but things don't always work out with cross-brand flashing, and if problems exist the last thing you want to do is send it in for RMA and your gskill drive comes up as an OCZ Vertex.

It would be interesting...would GSkill care that they got your money instead of OCZ but you went ahead and flashed the drive to OCZ's firmware so you could take advantage of OCZ's efforts with trim?

If I was GSkill I'd be all for it...I don't need to hire engineers to make a trim program work with my Falcon drives because the customer is willing to jump thru the hoops to get the OCZ version working.

I think the only entity that would not be happy here is OCZ as you are basically misappropriated (pirating by any other name...) their trim program and firmware to use on a drive they did not profit from you buying.

Of course if GSkill was really interested in torquing off their enthusiast base of customers they could exercise their rights to refuse warranty services on a cross-flashed drives, but that would be an interesting policy to enforce. They'd be in their rights, but it might not be to their benefit to exercise those rights.

 

[GarfieldtheCat]

I thought I read somewhere that the latest OCZ firmware is locked to their own drive, specifically because earlier firmware versions were being used in other competitors SSD drives