Seriously, though... what did you do last before this started?

Also... are you dealing with compressed or non compressed data? Compressed data will artificially bounce speeds up.

Are the "pre" numbers from before installing W8?

Also depends on your cache size and NAND. http://www.anandtech.com/show/6143/ocz-agility-4-256gb-review/3

How full was the drive at each instance?

(Seriously, though... what did you do last before this started?)

Installed a graphics card.

(Also... are you dealing with compressed or non compressed data? Compressed data will artificially bounce speeds up.)

Whatever the benchmark program uses, Doc, it was the same program and settings both times.

Are the "pre" numbers from before installing W8?

Windows 8 is on a different rig.  Windows 7 only.

That info is in the screenshots. 

Now: 39% of 238 GB.

Then:  23% of 238 GB.

Time for a yrag consult.

It will be interesting to hear, after you contact the maker of the SSD, as to what may have caused the difference in the write speeds. 

I will be researching the OCZ forums. I appreciate any and all input here from those who have SSDs or knowledge of.

Jim, I think your answer is going to be here:  http://www.anandtech.com/print/5817

and has to do with incompressible data write rates.

Here:  http://forums.anandtech.com/showthread.php?t=2281922

Anand's own testing showed rates comparable to what you're seeing, so I don't think you actually have a problem.

From what I have read, an SSD drive after it "settles in" loses some speed, even on WEI.

But half?

What's the point then?

It's still way above the speeds of a platter...

Sharpen your crayon.

Redneck i have owned MANY OCZ items and just RMA it they have excellent customer support even if you no longer have warranty they will do what they can to fix it for a small fee. 

part of the problem is OCZ drives are of very poor quality compared to much of what else is availabl despite all of the raving over them across tech. sites

sustained random write iops is the single most important metric across the board; try comparing a Mushkin "enhanced chronos" or current-generation Corsair drive off NewEgg or elsewhere to the current and recent-history generations of OCZ consumer drives...   there's a reason their company almost went under!

have you checked to see if any firmware updates are available?   ...did your benchmarking software get updated and begin disabling trim temporarily (or hopefully not permanently) for "testing purposes" ?

p.s; had a bit of trouble following this thread but did you say that you installed a new video card..? that may have had something to do with it - if you don't manually set your pagefile size windows would have automatically increased the initial paging filesize on your default page drive (boot drive, which I'm assuming is your ssd) while also increasing the reserved freespace for its expansion and copying any data from the front of the address-space to make room - that can be a taxing operation especially if your ssd had an abundance of temporary or unmovable files or/and pending updates around the time the operation occured

Hey RND,

to your screenshot it looks like something interfered while you made that diskmark...not the one you have currently made the old one
The thing is i tested a lot of SSD´s and never had a write speed that was higher than the reading spead on all of the write tests.
you should check out writing speed of your SSD....
i know it can go up to close 400 write im pretty sure it can lets say 350-60 but jumping over the marker 400 cant be correct since it is listed and tested with specifications  read: 400MB/s • write: 300MB/s so you can see there is no way it will cross the 400MB write marker...something has to be gone wrong in the previous test. The results of your current test are ok like Doc said and when comparing the test with the specs. .