First attempt at Kelemen style Metropolis back

(L) [2012/09/23] [ost by dylan] [First attempt at Kelemen style Metropolis] Wayback!

As well as the title, this is my first attempt at a path tracer.
My standard sampler code performs quite nicely, but my Metropolis sampler, which is copied nearly verbatim from Csaba's paper, is definitely wrong. It looks like old samples are contributing at new sample locations, or vice versa, and edges seem to be contrast enhanced, even wrapping around borders (note the bottom edge of the first image). Most of my test images seem to get darker (below the true value) as the sample count increases.

My not-quite-Metropolis sampler:
[LINK http://i.imgur.com/hbl8p.jpg]
[IMG #1 Image]
My path traced, noisy reference:
[LINK http://i.imgur.com/7vlzQ.jpg]
[IMG #2 Image]

Any ideas? I still can't see how my code is managing to clearly put samples at the wrong locations.

---

[IMG #1]:
[IMG #2]:

(L) [2012/10/04] [ost by dylan] [First attempt at Kelemen style Metropolis] Wayback!

The "stars" were a result of an uninitialised variable (well that's embarrassing). The contrast enhancement was due to a weighting problem, solved in [LINK http://igad2.nhtv.nl/ompf2/viewtopic.php?f=3&t=863&p=2221#p2221 this] thread.
Here's an updated fixed render:
[IMG #1 Image]

Does that refraction look right? I'm using an IOR of 1.5.

---

[IMG #1]:Not scraped: https://web.archive.org/web/20200925034336im_/http://imgur.com/rbN1x.jpg